V olumen 9 - PoliPapers

Transcription

Volumen 9 · publicación anual · 2014
Volumen 9
publicación anual · 2014
EDITORA
Mª Luisa Carrió Pastor
SECRETARIA
Mª José Labrador Piquer
COMITÉ EDITORIAL
Mª Dolores Forés López
Asunción Jaime Pastor
Mercedes López Santiago
Penny MacDonald
Rafael Seiz Ortiz
Debra Westall
COMITÉ CIENTÍFICO
Guadalupe Aguado de Cea (Universidad Politécnica de Madrid)
Francisco Alonso Almeida (Universidad de las Palmas de Gran Canaria)
Mª Antonieta Andión Herrero (Universidad Nacional de Educación a Distancia)
Reiner Arntz (Universität Hildesheim)
Paul Bangs (Freelance Consultant)
David Banks (Université de Bretagne Occidentale)
Marta Baralo (Universidad Antonio de Nebrija)
Françoise Blin (Dublin City University)
Pascual Cantos (Universidad de Murcia)
Mª Teresa Cabré (Universitat Pompeu Fabra)
Carmen Caldas-Coulthard (University of Birmingham)
Miguel Casas Gómez (Universidad de Cádiz)
Jozef Colpaert (Universidad de Amberes)
Malcolm Coulthard (University of Aston)
Raquel Criado (Universidad de Murcia)
Mª José Cuenca (Universitat de Valencia)
Graham Davies (ICT Consultant and Trainer - Modern Foreign Languages)
José Antonio Díaz Rojo (Consejo Superior de Investigaciones Científicas;
Instituto de Historia de la Ciencia y Documentación López Piñero, Valencia)
Mª Lluïsa Gea Valor (Universitat Jaume I)
Joan K. Hall (Penn State University)
Herbert J. Holzinger (Universitat de València)
Marta Iñigo Ros (Ecole de Traduction et Interpretation, Genève)
Michele Lazar (National University of Singapore)
Ana Mª Llinares García (Universidad Autónoma de Madrid)
Ricardo Mairal Usón (Universidad Nacional de Educación a Distancia)
Javier Martín Arista (Universidad de La Rioja)
Iliana Amalia Martínez (Universidad Nacional de Rio Cuarto, Argentina)
Gavin Melles (University of Melbourne)
Donna Rose Miller (University of Bologna)
Rafael Monroy Casas (Universidad de Murcia)
Carmen Muñoz Lahoz (Universidad de Barcelona)
Jo Anne Neff (Universidad Complutense de Madrid)
Carmen Pérez Vidal (Universidad Pompeu Fabra)
Germán Ruipérez García (Universidad Nacional de Educación a Distancia)
Francisco J. Ruiz de Mendoza Ibáñez (Universidad de La Rioja)
Bernd Rüschoff (Universität Duisburg-Essen)
Vicent Salvador i Liern (Universitat Jaume I)
Aquilino Sánchez (Universidad de Murcia)
Chris Taylor (University of Trieste)
Francisco Yus Ramos (Universidad de Alicante)
REVISTA INDEXADA EN:
ERIH European Reference Index for the Humanities: https://www2.esf.org/asp/ERIH/foreword/search.asp
DIALNET: http://dialnet.unirioja.es/appl/ppal.jsp
CINDOC: http://bddoc.csic.es:8080/index.jsp
Base de conocimientos de SFX (ExLibris)
METALIB: http://metalib.upv.es:3210/sfx_local/a-z?perform=textSearch
DOAJ: http://www.doaj.org
MLA: http://www.mla.org
Portal del hispanismo: Minsterio de Cultura. Instituto Cervantes: http://www.hispanismo.es/revista.asp?DOCN=2604
Linguistics Abstracts Online: http://www.linguisticsabstracts.com
Elektronische Zeitscherieftenbibliothez: http://www.zblx1.uni-regensburg.de/ezeit/
Latindex: http://www.latindex.unam.mx
JournalSeek: http://journalseek.net
DULCINEA: http://www.accesoabierto.net/dulcinea/consulta.php?directorio=dulcinea&campo=ID&texto=1647
EBSCO: http://www.ebsco.com/
Más información: ojs.upv.es/index.php/rdlyla
Departamento de Lingüística Aplicada
Universidad Politécnica de Valencia
Camino de Vera, s/n 46022 Valencia
DEPÓSITO LEGAL: V-1613-2006
ISSN: 1886-2438
© Departamento de Lingüística Aplicada
Diseño y maquetación
Triskelion Diseño Editorial
EDITA
Editorial UPV
ÍNDICE
1
Resolución de anáforas que requieren conocimiento cultural con la herramienta
FunGramKB
María de los Llanos Carrión Varela
Universidad Nacional de Educación a Distancia (Madrid)
14
Detection and correction of linguistic errors: results according to linguistic
preferences and uses
Jordi Cicres
Sílvia Llach
M. Dolors de Ribot
Universitat de Girona
23
Receptive vocabulary knowledge and motivation in CLIL and EFL
Almudena Fernández Fontecha
Universidad de La Rioja
33Approaching lexical loss in Canarian Spanish undergraduates: A preliminary
assessment
María Isabel González-Cruz
Universidad de Las Palmas de Gran Canaria
45The role of derivational paradigms with adjectival base in Old English
word-formation
Carmen Novo Urraca
55Alternations vs. variation in Old English. Methodological and descriptive issues
Ana Elvira Ojanguren López
67CAN LEXICAL ERRORS INFORM ABOUT WORD CLASS ACQUISITION IN THE FOREIGN LANGUAGE?:
EVIDENCE FROM GREEK LEARNERS OF SPANISH AS A FOREIGN LANGUAGE
Kiriakí Palapanidi
Universidad Abierta de Grecia
María Pilar Agustín Llach
Universidad de la Rioja
79Inventario fraseológico de las groserías en estudiantes de San Luis Potosí
Marco Antonio Pérez Durán
Oscar Arriaga Olguín
Universidad Autónoma de San Luis Potosí (México)
88The use of clickers to assess knowledge in foreign language classes and their
failure to increase reading compliance
Juan Pablo Rodríguez Prieto
Butler University, Indianapolis, United States
97HOW AN IDEA GERMINATES INTO A PROJECT OR THE INTRANSITIVE RESULTATIVE CONSTRUCTION
WITH ENTITY-SPECIFIC CHANGE-OF-STATE VERBS
Andreea Rosca
Centro Universitario de la Defensa (Zaragoza)
106Split and unified functions in the formation of Old English nouns and adjectives
Raquel Vea Escarza
RESEÑAS
117
Mancera, A. y Pano, A. (2013). El discurso político en Twitter: Análisis de mensajes que “trinan”. Barcelona:
Anthropos
Patricio Moya Muñoz.
Universidad Politécnica de Valencia.
RLyLA Vol. 09 (2014),
Revista de Lingüística y Lenguas Aplicadas
Vol. 9 año 2014, 1-13
EISSN 1886-6298
http://dx.doi.org/10.4995/rlyla.2014.2003
Resolución de anáforas que requieren conocimiento cultural con la
herramienta FunGramKB*
Universidad Nacional de Educación a Distancia (Madrid)
Abstract: La integración de diversos tipos de conocimiento lingüístico en sistemas de comprensión o procesamiento del
lenguaje natural (PLN) constituye una práctica común. Sin embargo, habitualmente se ha obviado la inclusión de conocimiento
cultural, ya sea por motivos técnicos o teóricos. No obstante, un sistema del PLN enriquecido con información cultural
constituye una herramienta más robusta y cohesionada, para llevar a cabo más óptimamente la resolución de problemas
vinculados a la comprensión del lenguaje natural como, por ejemplo, la ambigüedad existente en fenómenos lingüísticos como
la anáfora, referencia y correferencia o metáfora y metonimia, entre otros. El objetivo de este artículo es describir cómo la base
de conocimiento FunGramKB integra el conocimiento cultural a través de sus módulos conceptuales y, en especial, cómo la
información del módulo denominado Onomasticón puede contribuir a maximizar la informatividad del sistema completo, para
resolver problemas de ambigüedad en un determinado fenómeno lingüístico: la anáfora.
Palabras clave: PLN, FunGramKB, resolución de anáfora, bases de conocimiento, ingeniería del conocimiento.
Abstract: While integrating linguistic knowledge of any kind is becoming an almost implicit practice in natural language
understanding systems, the inclusion of cultural or world knowledge in these tools might have been neglected sometimes.
However, a NLP system or knowledge base enriched with cultural information is a more robust, better cohesioned instrument
for natural language understanding processes. The integration of this type of knowledge in NLP systems may be proven to
contribute to solving some phenomena that occur in natural language, such as anaphor, metaphor and metonymy, ambiguity
or co-reference, amongst others. The objective of this article is to describe the way FunGramKB (a knowledge base) integrates
cultural knowledge in its conceptual modules and, in particular, how the information contained in the Onomasticon module of
FunGramKB can contribute to maximising the informativeness and completeness of the whole system, thus resolving ambiguity
problems in a determined linguistic phenomenon: anaphora.
Keywords: NLP, FunGramKB, anaphora resolution, knowledge bases, knowledge engineering.
1. INTRODUCCIÓN
1.1. La ambigüedad en el lenguaje
Un hecho importante tras el cual subyacen muchos de los problemas y cuestiones por resolver en procesamiento
del lenguaje natural (en adelante, PLN) es que, para las computadoras, resulta una ardua tarea el poder entender el
lenguaje natural. Esta obvia característica, derivada de la ausencia de sentido común por parte de las máquinas, se
encuentra principalmente motivada o agravada por una característica innata del lenguaje natural: la ambigüedad.
Esta particularidad podría quizás ser consecuencia de la evolución natural favorecida por el principio de economía
en la lengua, mediante el cual el ser humano es capaz de transmitir el máximo de información posible utilizando
el mínimo de signos lingüísticos. Esta ley podría impulsar la existencia de ambigüedad lingüística de varios tipos:
ambigüedad semántica (de una misma lexicalización), ambigüedad sintáctica (provocada por elipsis y fenómenos
análogos), o incluso ambigüedades más allá del mero texto y que aparecen a nivel pragmático (por ejemplo,
utilizar una pregunta que se debe interpretar como una petición o incluso una orden).
* Este trabajo forma parte del proyecto de investigación FFI2011-29798-C02-01, financiado por el Ministerio de Ciencia e Innovación
Received: 2013/11/30 Accepted: 2014/02/15
|1
Resolución de anáforas que requieren conocimiento cultural con la herramienta FunGramKB
Por tanto, la resolución de las diversas manifestaciones de la ambigüedad en el lenguaje natural constituye uno
de los sempiternos campos de trabajo dentro del PLN, para lo cual intentaremos mostrar la aportación resolutoria
que la base de conocimiento FunGramKB puede ofrecer. En concreto, en el presente artículo se tratará de ilustrar
cómo la inclusión de conocimiento cultural en una base de conocimiento de semántica profunda como FunGramKB
puede ayudar a resolver algunos de los más frecuentes problemas de comprensión en el PLN y, específicamente,
uno de los fenómenos lingüísticos más comunes que guardan relación con la ambigüedad: la anáfora. Para ello,
primeramente se realizará una presentación de la herramienta en la que se desarrolla el actual trabajo: la base de
conocimiento FunGramKB, con un detalle más extenso del módulo dentro de la misma denominado Onomasticón.
Una vez definida la herramienta FunGramKB y su Onomasticón, en los diversos epígrafes de la sección 2 se
procederá a efectuar una contextualización de la problemática, mediante la descripción de diferentes fenómenos
anafóricos que son de interés para el PLN, así como la muestra de varios ejemplos citados en la literatura a los cuales
FunGramKB es capaz de aportar una nueva solución. Finalmente, la sección 3 se compone de las conclusiones y
futuros retos de investigación que se plantean a continuación del trabajo mostrado en el presente artículo.
1.2. Qué es FunGramKB
La herramienta FunGramKB1 (Periñán Pascual y Arcas Túnez 2004, 2007, 2008, 2010a, 2010b; Periñán Pascual
y Mairal Usón, 2009, 2010; Van Valin y Mairal Usón, en prensa; Mairal Usón, 2012; Mairal Usón et al. 2012; Periñán
Pascual y Mairal Usón, 2012) es una base de conocimiento léxico-conceptual multipropósito, creada para ser
implementada en diversas aplicaciones de PLN. La base FunGramKB es multifuncional, ya que ha sido diseñada
para ser utilizada en múltiples tareas de PLN, así como multilingüe, puesto que está soportada en varios idiomas2.
FunGramKB está estructurada en tres grandes niveles de información, que a su vez se subdividen en varios
módulos independientes pero interrelacionados (Periñán Pascual y Arcas Túnez, 2006; Periñán Pascual y Arcas
Túnez, 2010b). Estos tres grandes niveles son el nivel léxico (conocimiento lingüístico), el nivel gramatical
(conocimiento acerca de esquemas construccionales) y el nivel conceptual (conocimiento no lingüístico). La
integración de módulos y tareas permite a FunGramKB ser una herramienta versátil y completa, al combinar
diversos bloques que cubren tanto el aspecto lingüístico del lenguaje natural (módulos Léxico y Gramatical,
adaptados a cada una de las lenguas naturales soportadas) como el aspecto cognitivo (módulos Conceptuales),
que son los módulos compartidos por todas las lenguas naturales y que actúan como eje pivotal de la herramienta.
La figura que se incluye a continuación refleja la arquitectura de FunGramKB:
Figura 1. La arquitectura de FunGramKB3
1
2
3
http://www.fungramkb.com/
Alemán, búlgaro, catalán, español, francés, inglés e italiano.
Figura obtenida de http://www.fungramkb.com/ [fecha de consulta: 14/10/2013]
RLyLA Vol. 09 (2014), 1-13 | 2
Si se presta atención a los diferentes módulos lingüísticos, comenzando por el nivel léxico, se puede explicar
sus componentes de la siguiente forma: el Lexicón es el módulo que contiene la información morfosintáctica,
pragmática y colocacional de las unidades léxicas de cada lengua. Se trata, por tanto, de información que
ayuda a utilizar dichas unidades de manera correcta formalmente en el discurso. El Morficón de cada lengua
contiene aquellas reglas que afectan a la morfología flexiva, como por ejemplo las inflexiones verbales
o conjugaciones, así como las concordancias de género y número. El siguiente módulo lingüístico, el nivel
gramatical, contiene el denominado Constructicón, que almacena los Gramaticones de las diferentes lenguas
naturales, los cuales constituyen esquemas de construcciones que permiten construir la interfaz entre sintaxis
y semántica, denominada enlace (linking algorithm), para representar un texto de entrada en lenguaje natural
mediante una estructura lógica. Asimismo, en FunGramKB esta estructura lógica se ve realizada mediante un
nuevo formalismo denominado “estructura lógica conceptual” (CLS por sus siglas en inglés, Conceptual Logical
Structure), que maximiza la carga informativa y reduce la redundancia (Periñán Pascual y Arcas Túnez, 2010b).
El otro gran módulo que compone FunGramKB es el nivel conceptual. Este nivel almacena aquellas
representaciones prototípicas de la realidad que el individuo recrea en su mente, no las palabras con las que
se describe de manera lexicalizada dicha realidad. De este modo, mientras que el módulo de nivel léxico es
particular para cada lengua natural (ha de crearse, pues, un Lexicón para el inglés, otro para el español, y así
sucesivamente para cada lengua con cada uno de los tres componentes del nivel léxico), el nivel conceptual
es común a todos los lenguajes naturales, puesto que no está basado en conocimiento sobre las palabras
sino en conocimiento sobre el mundo (Periñán Pascual y Arcas Túnez, 2006). Es por esto que se puede
afirmar que el nivel conceptual de FunGramKB es universal, en tanto en cuanto los conceptos que recopila
y categoriza son comunes al mundo (Periñán Pascual y Arcas Túnez, 2007:199): “FunGramKB ontology takes
the form of a universal concept taxonomy, where ‘universal’ means that every concept we can imagine has an
appropriate place in this ontology.” Se trata de la lexicalización de estos conceptos, la manera de organizarlos
y de aglutinarlos en expresiones lingüísticas en las diferentes lenguas naturales, lo que difiere entre ellas. No
obstante, el hecho de que el módulo conceptual de FunGramKB sea denominado universal no lo exime de ser
lingüísticamente motivado, pero no lingüísticamente dependiente. Esto significa que cada uno de los conceptos
introducidos en el módulo conceptual (particularmente en la Ontología) tiene, necesariamente, al menos una
unidad léxica cuyo significado no coincide con ninguno de los otros conceptos ya presentes en la base de
conocimiento. Asimismo, este proceso asegura que cada nueva unidad léxica que pueda surgir en el futuro,
como consecuencia de la introducción de nuevas lenguas naturales en FunGramKB, tendría cabida dentro de
ella tras un proceso de negociación. La posibilidad de introducción de estos nuevos conceptos, lexicalizados
de una manera particular en una lengua natural, además, contribuye a eliminar una posible carga subjetiva
provocada por las propias lenguas maternas y contextos culturales de los ingenieros del conocimiento que
realizan la tarea de poblar la base (Periñán Pascual y Arcas Túnez, 2010b).
El nivel conceptual de FunGramKB se compone de tres módulos: la Ontología, que es una estructura
jerárquica de conceptos usados por las personas para describir cualquier situación cotidiana; el Cognicón, que
almacena conocimiento procedimental en forma de secuencias temporales o guiones, y que está basado en el
modelo temporal de Allen (Allen, 1983; Allen y Ferguson, 1994); y el Onomasticón, que almacena conocimiento
episódico y cultural acerca de entidades (p. ej., personas, ciudades, lugares, acontecimientos, etc.) en forma
de bio-estructuras.
A continuación, describiremos brevemente los tres módulos del nivel conceptual de FunGramKB, puesto que
éste será el módulo en el que se centrarán las tareas que mostramos en el presente trabajo, más concretamente
en el Onomasticón.
La Ontología es el elemento central de FunGramKB. Comprende tres tipos generales de conceptos (llamados
metaconceptos y señalados con el símbolo (#)): entidades, (# ENTITY), eventos (# EVENT) y cualidades (#
QUALITY). Estos tres conceptos organizan la dimensión cognitiva de los nombres, verbos y adjetivos,
respectivamente (Periñán Pascual y Arcas Túnez, 2007). Estas tres dimensiones conceptuales se relacionan
entre sí a través de los llamados postulados de significado (Meaning Postulates, MPs), que son constructos
conceptuales que definen la realidad. El siguiente nivel bajo los metaconceptos se compone de los conceptos
básicos, precedidos del símbolo (+), y por debajo se encuentran los conceptos terminales, identificados
mediante el símbolo ($), teniendo todo ellos un sufijo numérico al final (_00, _01, etc.) . A modo de aclaración,
podemos observar la figura que ilustra la esta jerarquía de unidades conceptuales en Periñán Pascual y Arcas
Túnez (2007a) para un ejemplo concreto, el concepto terminal $FOOTBALL_00:
RLyLA Vol. 09 (2014), 1-13 | 3
Figura 2. Ejemplo de estructuración ontológica en FunGramKB (Periñán Pascual y Arcas Túnez, 2007:199).
El Cognicón es el módulo que almacena conocimiento procedimental en forma de secuencias temporales o
guiones de situaciones cotidianas (por ejemplo, ir al cine, visitar un museo, ir a un restaurante, etc.), y que está
basado particularmente en el modelo temporal de Allen (Allen, 1983; Allen y Ferguson, 1994). En FunGramKB, se
estructura como una serie de predicaciones dentro de un marco lineal temporal. Un guión, por tanto, comprende
varias predicaciones, y cada una de ellas se entiende como un evento E, tratado como un intervalo entre un par
de puntos temporales: el punto temporal i, que es el inicio de la acción, y el punto temporal t, que es su fin. Así
como Allen había previsto la posibilidad de diversas relaciones entre estos eventos (simultaneidad, posterioridad
o solapamiento, entre otros), también quedan éstas reflejadas en FunGramKB (Periñán Pascual y Arcas Túnez,
2010b; Periñán Pascual, 2012).
Finalmente, el Onomasticón está formado por aquellas entidades que componen nuestro conocimiento
enciclopédico o histórico, por ejemplo personas, lugares, acontecimientos, etc., en forma de bio-estructuras. La
naturaleza y características de este módulo serán desarrolladas manera algo más extensa en la siguiente sección,
con objeto de poder mostrar más adelante cómo el conocimiento cultural almacenado en el Onomasticón de
FunGramKB sirve como apoyo para resolver ejemplos concretos de fenómenos anafóricos, que requieren algo
más que mero conocimiento lingüístico (léxico o sintáctico) para ser desambiguados con éxito.
1.3. El Onomasticón de FunGramKB
El Onomasticón de FunGramKB está encuadrado dentro del nivel conceptual de la base de conocimiento, junto
a la Ontología y al Cognicón. El Onomasticón, como se ha reseñado con anterioridad, almacena la información
relativa a las instancias de entidades y eventos, en forma de bio-estructuras. Con la finalidad de comprender qué
es a lo que se apunta de manera precisa mediante la expresión bio-estructura, es necesario mencionar el carácter
prototípico del conocimiento almacenado en los módulos conceptuales que son la Ontología y el Cognicón.
Dada la prototipicidad que poseen, llamamos a las estructuras que forman la Ontología y el Cognicón protoestructuras, puesto que se deben a la generalidad de un concepto y no a la particularidad. No obstante, a pesar
de reflejar esta generalidad, en el marco de una situación prototípica, cuando cabe la posibilidad de que alguna
de sus características pueda variar en algún momento hipotético del tiempo, se utiliza la etiqueta de rebatible, lo
que permitirá una herencia denominada no-monotónica. Por ejemplo, si hablamos de un pájaro, una característica
prototípica de este tipo de animales es “puede volar” pero, puesto que es posible que exista un pájaro que, a
pesar de serlo, no sea capaz de volar (p.ej. un pingüino o un avestruz), el rasgo “puede volar” será entonces una
característica rebatible.
Por el contrario, el Onomasticón refleja la situación opuesta, donde la prototipicidad deja paso a la especificidad,
y puesto que se trata de entidades reales existentes, se utiliza la etiqueta de bio para nombrar a estas estructuras.
RLyLA Vol. 09 (2014), 1-13 | 4
De este modo, mientras que un concepto de la Ontología es +SONG_00, una entidad del Onomasticón es
%HEY_JUDE_004.
Además del parámetro de la prototipicidad, los esquemas conceptuales en FunGramKB también se clasifican
conforme a otro parámetro: la temporalidad (Periñán Pascual y Arcas Túnez, 2010b). Teniendo en cuenta esta
característica, el conocimiento dentro de los esquemas conceptuales puede presentarse de manera temporal o
atemporal. Si se hace de manera temporal, significará que ese conocimiento se presenta dentro de un marco de
tiempo (lo que se denominan macroestructuras). Por ejemplo, la biografía de una persona (una persona que sea una
entidad del Onomasticón) o un guión de los reflejados en el Cognicón poseerían la propiedad de ser estructuras
temporales. Por otro lado, cabe la posibilidad de que las estructuras conceptuales representen el conocimiento
de manera atemporal (microestructuras), como sucedería con una característica aislada de una entidad del
Onomasticón o de la Ontología. La convergencia de estos dos parámetros, prototipicidad y temporalidad, resulta
en la creación de una tipología de estructuras conceptuales compuesta por proto-microestructuras, protomacroestructuras, bio-microestructuras y bio-macroestructuras. Si se coloca en una matriz esta combinación de
parámetros, se obtendrá la siguiente tabla que lo ilustra:
PROTOTIPICIDAD
TEMPORALIDAD
+
-
-
+
Proto-microestructura
(Postulado de significado)
Proto-macroestructura
(Guión)
Bio-microestructura
(Retrato)
Bio-macroestructura
(Historia)
Tabla 1. Tipología de esquemas conceptuales en FunGramKB (Periñán Pascual y Arcas Túnez, 2010b).
Además de estar organizadas de la manera arriba ilustrada, estas cuatro estructuras se integran en la base
FunGramKB, lo que puede observarse representado de manera gráfica en la figura siguiente:
COGNICÓN
ONTOLOGÍA
PROTO-MICROESTRUCTURAS
(POSTULADOS DE SIGNIFICADO)
PROTO-MACROESTRUCTURAS
(GUIONES)
COREL
BIO–ESTRUCTURAS
BIO-MICROESTRUCTURAS
(RETRATOS)
BIO-MACROESTRUCTURAS
(HISTORIAS)
ONOMASTICÓN
Figura 3. El Planeta Cognitivo (Periñán Pascual y Mairal Usón, 2010:15)
4
Nótese el símbolo (%) utilizado para identificar las entidades del Onomasticón, diferente de los símbolos (#), (+) o ($), que identifican a las entidades pertenecientes a
la Ontología como metaconceptos, conceptos básicos o conceptos terminales, respectivamente.
RLyLA Vol. 09 (2014), 1-13 | 5
Esta figura muestra la necesaria interrelación entre los tres módulos del nivel conceptual de FunGramKB a través
de un mismo lenguaje de representación, denominado COREL5, ya que, al igual que ocurre con los componentes
de la memoria humana a largo plazo, un sistema de PLN que persiga permitir con éxito el razonamiento ha de
prever que los componentes que lo forman puedan relacionarse entre ellos a través de un mismo lenguaje (Periñán
Pascual y Arcas Túnez, 2010b; Periñán Pascual y Mairal Usón, 2010).
La Tabla 1 y la Figura 3 arriba mostradas sirven de contexto a la hora de describir el Onomasticón y los elementos
que éste alberga. Así pues, si observamos una entidad como, por ejemplo, Elvis Presley (%ELVIS_PRESLEY_00,
en notación COREL), dependiendo del tipo de característica que se desee reflejar, se necesitará utilizar un tipo
u otro de estructura conceptual. Por tanto, si deseamos describir la profesión de cantante de Elvis, será preciso
emplear una bio-microestructura, ya que dicha propiedad puede encuadrarse como uno de los muchos rasgos
que, de manera relativamente aislada del resto de características, describen a Elvis en un momento concreto del
tiempo, como serían, entre otras muchas, la profesión de actor, su color de pelo o su estatura (cf. Periñán Pascual
y Arcas Túnez, 2010b). Es por ello que este tipo de microestructuras en el Onomasticón se denominan “retratos”
(snapshots en su versión en inglés), por el hecho de asemejarse a una fotografía tomada en un momento particular.
Sin embargo, si se desea describir la biografía de Elvis, es necesario hacer uso de una bio-macroestructura, las
llamadas “historias” (stories) por obvias razones: contienen elementos que han de ser integrados dentro de un
esquema temporal determinado (p. ej., la fecha de nacimiento o fallecimiento del cantante), ya que se trata de
elementos cuya existencia se encuadra en un marco temporal establecido y cuyo orden no es posible alterar.
Otro ejemplo al que se puede aludir es una entidad como el Taj Majal. De nuevo, si perseguimos describir una
propiedad como “hecho de mármol” o “mausoleo”, se haría mediante una bio-microestructura (retrato), mientras
que el relato de su proceso de construcción habría de realizarse mediante una historia (bio-macroestructura).
Esto se puede ilustrar en lenguaje COREL a través del siguiente ejemplo, originalmente sugerido en Periñán
Pascual y Arcas Túnez (2010b) acerca de la entidad “%TAH_MAHAL_00”. El equivalente en lenguaje natural de las
predicaciones representadas se muestra a continuación de cada una de ellas:
+(e1: +BE_02 (x1: %TAH_MAHAL_00)Theme (x2: %INDIA_00)Location)
El Taj Majal está ubicado en India.
*(e2: +BE_01 (x1)Theme (x3: +WHITE_00 & $MARBLE_00)Attribute)
Su principal material es el mármol blanco.
*(e3: +COMPRISE_00 (x1)Theme (x4: 1 $DOME_00 & 4+TOWER_00)Referent)
El Taj Majal tiene una cúpula principal y cuatro torres.
Estas tres predicaciones acerca del Taj Majal pertenecen a la categoría “retrato”, ya que las tres enumeran
rasgos atemporales de la entidad. Sin embargo, apreciamos que sólo la primera de ellas (e1), porta el símbolo (+)
al frente, lo que indica que tan sólo esa predicación es estricta. Las otras dos predicaciones son rebatibles, lo que
se indica en lenguaje COREL mediante el símbolo (*), ya que si el Taj Majal encontrara destruida una de sus torres
o la cúpula principal por cualquier circunstancia, o si alguna de éstas se sustituyera por un elemento hecho de
otro material diferente del mármol blanco, el Taj Majal seguiría considerándose la misma entidad. No obstante, si
el Taj Majal no se encontrase en la India, entonces ya no se trataría de la misma entidad, sino de una reproducción
del mausoleo en otra ubicación.
Por otro lado, y continuando con el mismo ejemplo, es posible también representar características temporales
del Taj Majal. Es el caso de las siguientes predicaciones y su equivalente en lenguaje natural a continuación (cf.
Periñán Pascual y Arcas Túnez, 2010b):
+(e1: past +BUILD_00 (x1)Theme (x2: %TAH_MAHAL_00)Referent (f1: 1633)Time)
El Taj Majal fue construido en 1633.
+(e2: past +BE_00 (x2)Theme (x3:%WORLD_HERITAGE_SITE_00)Referent (f2: 1983)Time)
El Taj Majal se convirtió en Patrimonio de la Humanidad por la UNESCO en 1983.
Los conceptos representados arriba constituyen historias, ya que, además de ser predicaciones estrictas,
indicado por el signo (+) frente a cada una de ellas (no es posible cambiar eventos sucedidos en el pasado, como
el año de construcción, ni aquel en el que fue declarado Patrimonio de la Humanidad), pertenecen a un esquema
temporal en el que están encuadradas de una manera precisa y determinada. Por consiguiente, se observa a
través de los ejemplos anteriormente mencionados cómo una misma entidad puede compartir retratos e historias
que representen rasgos conceptuales que la definan.
5
Conceptual Representation Language, interlengua utilizada como lenguaje de representación a través de todos los módulos conceptuales de FunGramKB.
RLyLA Vol. 09 (2014), 1-13 | 6
Otro de los principales rasgos distintivos del Onomasticón frente a Ontología y Cognicón es que la manera de
poblar este módulo no es enteramente manual, como sucede en los otros dos módulos, sino que se realiza de
manera semi-automática, cuyo procedimiento está descrito en Periñán Pascual y Carrión Varela (2011). Mediante
esta metodología semi-automática se posibilita que, tras la elaboración manual de plantillas de proyección
(denominadas reglas), la inserción de éstas en la herramienta FunGramKB permita el trasvase automático de
información a FunGramKB procedente de otras fuentes, como es principalmente DBpedia6 (Auer et al., 2007;
Bizer et al., 2009). Este procedimiento de población semi-automática permite que la gran profusión de datos
contenida en DBpedia, cuyo principal origen de importación de datos es Wikipedia7, sea transferida a FunGramKB
y actualizada a la vez que lo hacen las fuentes que la alimentan. La principal fuente de información que DBpedia
toma de Wikipedia para poblar su base de conocimiento son los denominados info-boxes, elementos incluidos en
los artículos de Wikipedia y cuyo ejemplo se ilustra a continuación:
Ilustración 1. Info-box de Wikipedia (Periñán Pascual y Carrión Varela, 2011:93).
Mediante la inclusión en FunGramKB de los contenidos que cada info-box proporciona para cada entidad,
se compone una estructura conceptual más completa y que permite resolver ejemplos de fenómenos anafóricos
complejos, como se describirá a continuación.
2. LA ANÁFORA
Uno de los fenómenos lingüísticos en los que más se ha trabajado a nivel de PLN y cuya resolución ha sido
buscada por numerosos proyectos (p. ej., en Mitkov, 2002 se puede encontrar un exhaustivo repaso a dicho
propósito a través de la historia) es la anáfora.
La anáfora es una forma de presuposición que consiste en apuntar a un referente o elemento previo mencionado
(Halliday y Hasan, 1976), lo que contribuye en gran manera a la cohesión de un texto, especialmente la llamada
anáfora endofórica (cuyo antecedente se encuentra dentro del mismo texto), ya que permite la continuación lógica
y coherente de una secuencia textual, en lengua escrita u oral, sin necesidad de reiterar continuamente las mismas
expresiones lingüísticas o lexicalizaciones para aludir a un mismo concepto o referente. Según resumen Fan et al.
(2005), existen tres elementos clave en la construcción de una anáfora: el antecedente o anclaje, la expresión que
se refiere al antecedente y el enlace o link:
The object that is being referred to is called the anchor or the antecedent, the expression that refers to the antecedent
is called the referring expression, and the association between the referring expression and the anchor is called the
link. (Fan et al., 2005: 153).
6
7
http://dbpedia.org/About
http://en.wikipedia.org/wiki/Main_Page
RLyLA Vol. 09 (2014), 1-13 | 7
Para la ilustración de este fenómeno en puridad, así como para mostrar también el tratamiento y resolución
que FunGramKB efectúa de otros fenómenos lingüísticos relacionados, en este artículo nos referiremos al término
anáfora en su concepción más amplia. Esto incluye a otros fenómenos relacionados como, por ejemplo, la
catáfora, puesto que el objetivo principal de los ejemplos y descripciones presentados más adelante es vincular las
cadenas de correferencia existentes en el lenguaje natural, tanto las que hacen alusión a elementos previamente
mencionados, como sucede en la anáfora, así como aquellas cadenas correferenciales materializadas mediante la
mención posterior de la entidad nombrada, lo que conocemos como catáfora.
2.1. Resolución de antecedentes anafóricos de entidades nombradas
Si bien para la mayoría de los hablantes nativos de una lengua natural es algo relativamente sencillo el
identificar y resolver el antecedente al que hace alusión una expresión anafórica, no siempre sucede de este
modo para las máquinas que procesan el lenguaje natural. Con objeto de ilustrar este fenómeno, observemos los
siguientes ejemplos en los cuales se han destacado las alusiones a entidades nombradas (antecedentes) y sus
correspondientes anáforas:
(i)
Good morning from Hollywood. You know, here at the famous Grauman’s Chinese Theater, they have handprints
and footprints from all sorts of famous people, from Jimmy Stewart to the cast of Harry Potter. And this will be
the latest, Michael Jackson, the imprint of his famous sequined glove over there, his footprints. And then the
handprints of his three children who, just hours ago, took steps to ensure their father’s Hollywood immortality.8
(ii) The Nation has since denied any “official business or professional relationship” with Jackson, and yet several
of their members, sober-faced and in business suits, lined up outside the Santa Maria courthouse at the
singer’s arraignment last January. More than a thousand Jackson supporters thronged the streets in front of the
courthouse, holding up signs and cheering as they strained for a glimpse of Michael, who was accompanied that
morning by his mother, father, brother Jermaine and sister Janet. Suddenly a roar went up. There was the King
of Pop on the roof of a vehicle waving his arms and blowing kisses at the screaming crowd.9
En (i), se observa que la referencia al fallecido cantante Michael Jackson se replica mediante la utilización en
tres ocasiones del determinante posesivo his, lo que facilita la cohesión textual y permite identificar, aun en el
supuesto caso de ignorar quién es Michael Jackson, que his se refiere en las tres ocasiones al mismo referente o
entidad masculina. Además, la mención posterior como their father se debe unir al referente his three children, que
a nivel conceptual o semántico queda vinculado a their father, lo que hace posible interpretar el texto de manera
coherente por parte del receptor del mensaje.
En caso de ser interpretado por una máquina, concretamente por la herramienta FunGramKB, la interpretación
de las expresiones his en relación a their father contenidas en (i) podría ser resuelta sin mayor problemática,
gracias a la carga conceptual y morfológica que poseen los diferentes módulos de la base de conocimiento y que
permite, por medio de su razonador, vincular los conceptos entre sí10.
Sin embargo, para identificar el sexo de la entidad Michael Jackson y, de este modo, el correspondiente
género gramatical que permitiría la vinculación anafórica entre la entidad nombrada, el determinante posesivo his
y el posterior their father, resulta de gran importancia el módulo Onomasticón de FunGramKB. En este ejemplo
concreto, gracias a la importación de la información contenida en Wikipedia, es posible saber que la entidad
Michael Jackson es un cantante de sexo masculino, como se ilustra a continuación:
8
9
10
For January 27, 2012, CBS 2012 (120127). CBS_ThisMorning. Obtenido a través del corpus COCA, Davies (2008-).
Cheo Hodari Coker, MAN IN THE MIRROR, 2004 (Apr) Vol. 34, Iss. 12; pg. 186, 4 pgs, Essence. Obtenido a través del corpus COCA, Davies (2008-).
Por ejemplo, gracias al conocimiento conceptual almacenado en la Ontología, el razonador de FunGramKB sabe que un padre es una persona de género masculino
que tiene descendencia.
RLyLA Vol. 09 (2014), 1-13 | 8
Ilustración 2. La entidad “Michael Jackson” en Wikipedia11
En relación al ejemplo (ii), éste pone de manifiesto de manera más explícita la necesidad e importancia de
la existencia de un módulo en FunGramKB que se ocupe de nutrir la base de conocimiento con información
acerca de entidades nombradas. De otro modo, la alusión a King of Pop que aparece en dicho texto no podría
ser resuelta de manera exitosa por parte de la máquina, la cual, basándose tan sólo en coincidencias léxicogramaticales, tendría dificultades en determinar qué entidad nombrada masculina mencionada anteriormente en el
texto (father, brother Jermaine, Michael) desempeña el papel de antecedente de la expresión King of Pop. Gracias
a la información cultural y enciclopédica incluida en el Onomasticón e importada de Wikipedia, sabemos que King
of Pop se refiere a Michael Jackson, como evidencia el artículo de Wikipedia mostrado en la Ilustración 2.
Otro ejemplo que mostraría la capacidad de FunGramKB para desambiguar cadenas correferenciales lo
podemos obtener de la obra de Mitkov (2002), donde se menciona una cadena correferencial (expresiones que
tienen el mismo referente a lo largo del mismo extracto textual):
(iii) Sophia Loren says she will always be grateful to Bono. The actress revealed that the U2 singer helped her calm
down when she became scared by a thunderstorm while travelling on a plane. (Mitkov, 2002:5).
En (iii) podemos observar cómo se efectúa una cadena de correferencias entre Sophia Loren, she y the actress,
tal como destaca el autor mediante el empleo de la letra cursiva. Por otro lado, encontramos que entre “Bono”
y “U2 singer” también existe una correferencia. Sin embargo, en caso de que una máquina tuviera que decidir
cuáles son las cadenas de correferencia en este extracto, podría existir ambigüedad a la hora de decidir si the
actress hace referencia a Sofia Loren o a “Bono”, al igual que la mención a “the U2 singer”. No obstante, puesto
que dicha información se encuentra dentro de los datos que Wikipedia importa al Onomasticón de FunGramKB,
de nuevo quedaría salvaguardada la correcta interpretación referencial.
2.2. Resolución de la anáfora indirecta
En ocasiones, la coherencia de un extracto textual va más allá de una mera cadena correferencial existente
en la misma oración o extracto, para desplazarse a referentes más alejados, incluso más allá del emisor, el
receptor o los elementos contextuales del momento de la emisión del mensaje. Este fenómeno es lo que se
denomina anáfora indirecta. Según Mitkov (2002:15), “Indirect anaphora arises when a reference becomes part
of the hearer’s or reader’s knowledge indirectly rather than by direct mention”, lo que significa que la conexión
de una expresión anafórica con su antecedente resulta más complicada, al no estar basada en una identificación
sintácticamente detectable, sino que va más allá, precisando de conocimiento general del mundo. Otros autores
se refieren a fenómenos relacionados mediante diferentes denominaciones, como por ejemplo Eckert y Strube
(citados en Palomar et al., 2000), que lo denominan anáfora abstracta, si bien es cierto que este término lo aplican
de una manera más general a cualquier antecedente que no sea un sintagma nominal (Palomar et. al, 2000: 206):
“According to Eckert and Strube (1999), if the antecedent is a noun phrase then the anaphora is classified as
individual anaphora, otherwise, the anaphora is classified as abstract anaphora.” Un nuevo término que podemos
relacionar también es el concepto de bridging (Palomar et al., 2000; Fan et al., 2005), referido a las relaciones
11
http://en.wikipedia.org/wiki/Michael_Jackson [fecha de consulta 15/10/2013].
RLyLA Vol. 09 (2014), 1-13 | 9
(excluyendo la de identidad) que existen entre una anáfora y su antecedente, lo que nos permite encuadrar aquí
relaciones como la hiponimia o hiperonimia:
Clark (1977) called bridging descriptions to definite descriptions that either have an antecedent denoting the same
discourse entity, but using a different head noun (synonym, hypernym or hyponym) or are related by other relation
than identity. (Palomar et al., 2000: 207).
Debido a la gran complejidad que entraña la resolución de la anáfora indirecta, diversos autores (Mitkov, 2002;
Fan et al., 2005) concluyen a este respecto lo que se resume en Mitkov (2002:34): “Therefore anaphors requiring
real-world knowledge for their resolution stand the least chance of being resolved successfully”, lo que parece
confirmar que, si en muchos casos este tipo de anáforas pueden resultar complicadas de resolver incluso para
un humano que no posea ciertos conocimientos culturales, la resolución por parte de la máquina puede tornarse
una quimera. Precisamente por esto, es en estos casos donde el Onomasticón de FunGramKB puede resultar
de especial ayuda para conferir sentido a las referencias expresadas, ya que la integración de conocimiento
enciclopédico en este módulo supone un enriquecimiento cultural de la herramienta, lo que permite la resolución
de casos de anáfora indirecta, como veremos a continuación.
Observemos el siguiente ejemplo:
(iv)When Take That broke up, the critics gave Robbie Williams no chance of success. (Mitkov, 2002:15).
En este caso, la información acerca de lo que es la entidad Take That es necesaria para comprender dónde
está la correlación o coherencia semántica de aludir a otra entidad no mencionada previamente, Robbie Williams,
ya que a simple vista no parece establecerse ninguna cadena correferencial. Sin embargo, la motivación de
mencionar a la entidad Robbie Williams en este ejemplo responde a su condición de integrante de la banda
musical inglesa Take That, una información que Wikipedia (y, por ende, el Onomasticón de FunGramKB) sí es
capaz de aportar acerca de la banda y sus integrantes:
Ilustración 3. La entidad “Take That” en Wikipedia12
Otra instancia más compleja de cómo FunGramKB es capaz de resolver este problema se puede observar en
el ejemplo que mostramos en (v). Para poder proporcionar una respuesta exitosa a la desambiguación anafórica
del extracto, es necesario que la herramienta posea información relativa a eventos de corte histórico, lo que va
más allá de un conocimiento básico de entidades nombradas individuales, para referirse a eventos complejos de
la historia, conocidos por los hablantes medios de una comunidad pero ignorados, en su gran mayoría, por las
máquinas. Observemos la siguiente expresión destacada:
(v) It is possible that some human should have climbed the World Trade Center Towers without ropes before they
were destroyed. (King, 2013).
12
http://en.wikipedia.org/wiki/Take_That [fecha de consulta 21/10/2013].
RLyLA Vol. 09 (2014), 1-13 | 10
En este caso, para que la máquina sea capaz de efectuar la resolución anafórica del antecedente del pronombre
they, es necesario que ésta posea conocimiento acerca de que las torres del World Trade Center fueron destruidas
en una fecha precisa, ya que los humanos a los que se menciona en el mismo ejemplo, tanto sintáctica como
conceptualmente, podrían ser un potencial antecedente13. No obstante, el Onomasticón proporciona el vínculo
conceptual con el conocimiento enciclopédico necesario para desambiguar el antecedente del pronombre
they, como se ilustra en la Figura 3, que muestra una representación gráfica de la conexión conceptual entre la
información del info-box de Wikipedia para la entidad “World Trade Center Towers”14, la fecha de destrucción de
dicha entidad y la propiedad “destructionDate” del Onomasticón:
Figura 4. Vínculo conceptual de las propiedades de la entidad “World Trade Center Towers” en Wikipedia
y el Onomasticón de FunGramKB.
En la Figura 4 observamos la siguiente conexión conceptual: la información que Wikipedia aporta acerca
del hecho histórico de la destrucción de las torres del World Trade Center se basa tanto en la categorización de
estado de la entidad (“Status: Destroyed”) como en la asignación de una fecha concreta a dicho acontecimiento
(“September 11, 2001”). La conexión de esta información con FunGramKB la encontramos a través del
Onomasticón, a través de la propiedad “destructionDate”. Esta propiedad determina la fecha en la que cualquier
entidad catalogada como “Building” (la entidad World Trade Center Towers lo es) ha sido destruida (si lo ha
sido). Por tanto, este vínculo conceptual permite que, en el ejemplo (v), se resuelva de una manera más clara y
menos ambigua que el antecedente del pronombre they es la entidad World Trade Center Towers, lo que pone de
manifiesto la capacidad de la base de conocimiento FunGramKB para actuar de apoyo a la hora de resolver de
manera más exitosa cuestiones y problemas derivados del PLN.
3. CONCLUSIONES
En el presente artículo se ha tratado de mostrar cómo FunGramKB y la inclusión de conocimiento cultural
a través del Onomasticón posibilitan la ampliación del potencial de dicha herramienta, con objeto de ayudar a
desambiguar fenómenos anafóricos en los que se requiere conocimiento cultural o enciclopédico. Puesto que
la concepción de FunGramKB como herramienta multipropósito es una premisa sobre la cual se basa nuestra
labor de investigación presente y futura, en los trabajos actualmente en desarrollo se incluye la aplicación del
Onomasticón de FunGramKB a la resolución de otros problemas lingüísticos que acucian a los sistemas de PLN,
como son la referencia y correferencia, metáfora y metonimia. Dicho trabajo encontrará su publicación en la tesis
doctoral actualmente en elaboración por parte de la autora del presente artículo.
13
14
Puesto que los humanos, según la información conceptual contenida en la Ontología de FunGramKB, son entidades que también pueden ser destruidas.
http://en.wikipedia.org/wiki/World_Trade_Center_Towers, fecha de consulta [23/10/2013].
RLyLA Vol. 09 (2014), 1-13 | 11
REFERENCIAS
Allen, J.F. (1983). “Maintaining knowledge about temporal intervals”, Communications of the ACM, 26/11: 832843. doi:10.1145/182.358434
Allen, J.F. y Ferguson, G. (1994). “Actions and events in temporal logic”, Journal of Logic and Computation, 4/5:
531- 579. doi:10.1093/logcom/4.5.531
Auer, S., Bizer, C., Lehmann, J., Kobilarov, G., Cyganiak, R., e Ives, Z. (2007). “DBpedia: A Nucleus for a Web of
Open Data”, en Aberer et al. (Eds.), The Semantic Web, 6th International Semantic Web Conference, 2nd
Asian Semantic Web Conference, ISWC 2007 + ASWC 2007, Busan, Korea, November 11–15, 2007. Berlin/
Heidelberg: Springer. Disponible en http://www.informatik.uni-leipzig.de/~auer/publication/dbpedia.pdf
[fecha de consulta: 14/10/2013].
Bizer, C., Lehmann, J., Kobilarov, G., Auer, S., Becker, C., Cyganiak, R. y Hellmann, S. (2009). “DBpedia –
A Crystallization Point for the Web of Data”, Journal of Web Semantics: Science, Services and Agents on the
World Wide Web, Issue 7: 154-165. Disponible en http://jens-lehmann.org/files/2009/dbpedia_jws.pdf
[fecha de consulta: 14/10/2013].
Davies, M. (2008-). The Corpus of Contemporary American English: 450 million words, 1990-present. Disponible
en http://corpus.byu.edu/coca [fecha de consulta: 13/11/2013].
Fan, J., Barker, K. y Porter, B. (2005). “Indirect anaphora resolution as semantic path search”, en Proceedings of
the 3rd international conference on Knowledge capture (K-CAP ‘05). ACM, New York, NY, USA, pp. 153160. Disponible en: http://doi.acm.org/10.1145/1088622.1088650 [fecha de consulta: 28/11/2013].
Halliday, M.A.K. y Hasan, R. (1976). Cohesion in English. London: Longman.
King, J.C. (2013). Anaphora, “The Stanford Encyclopedia of Philosophy” (Summer 2013 Edition), Edward N.
Zalta (ed.), Disponible en: http://plato.stanford.edu/archives/sum2013/entries/anaphora/
[fecha de
consulta: 23/10/2013].
Mairal Usón, R. (2012). “La arquitectura de una base de conocimiento léxico conceptual: implicaciones lingüísticas”,
en M. Giammatteo, L. Ferrari y H. Albano (eds.). Léxico y Sintaxis. Volumen temático de la serie editada por
la Sociedad Argentina de Lingüística. Mendoza: Editorial FFyL, UNCuyo y SAL, pp. 183-210. Disponible en
http://ffyl.uncu.edu.ar/spip.php?article3638 [fecha de consulta: 13/11/2013].
Mairal Usón, R., Periñán Pascual, C. y Pérez Cabello de Alba, M. B. (2012). “La representación léxica. Hacia un
enfoque ontológico”, en R. Mairal Usón, L. Guerrero y C. González (eds.) El funcionalismo en la teoría
lingüística. La Gramática del Papel y la Referencia. Introducción, avances y aplicaciones. Madrid: Akal,
pp. 85-102.
Mitkov, R. (2002). Anaphora Resolution. Pearson Education Limited. Great Britain: Longman.
Palomar Sanz, M., Saiz Noeda, M., Muñoz Guillena, R., Suárez Cueto, A. y Martínez Barco, P. (2000). “PHORA: a
system to solve the anaphora in Spanish”, en Proceedings of Third International Conference on Discourse
Anaphora and Anaphor Resolution (DAARC2000). Lancaster: University of Lancaster, 2000, pp. 206-211.
Periñán Pascual, C. (2012). “The situated common-sense knowledge in FunGramKB”, Review of Cognitive
Linguistics, 10/1: 184-214. doi:10.1075/rcl.10.1.06per
Periñán Pascual, C. y Arcas Túnez, F. (2004). “Meaning postulates in a lexico-conceptual knowledge base”, en
Proceedings of the 15th International Workshop on Databases and Expert Systems Applications. Los
Alamitos (California): IEEE, pp. 38-42.
Periñán Pascual, C. y Arcas Túnez, F. (2006). “Reusing Computer-oriented Lexica as Foreign-Language Electronic
Dictionaries”, Anglogermanica online: Revista electrónica periódica de filología alemana e inglesa, nº 4,
pp. 69-93. Disponible en: http://anglogermanica.uv.es:8080/Journal/Viewer.aspx?Year=2006&ID=perinan.
pdf [fecha de consulta: 14/10/2013].
Periñán Pascual, C. y Arcas Túnez, F. (2007). “Cognitive modules of an NLP knowledge base for language
understanding”, Procesamiento del Lenguaje Natural, 39: 197-204.
Periñán Pascual, C. y Arcas Túnez, F. (2008). “A cognitive approach to qualities for NLP”, Procesamiento del
Lenguaje Natural, 41: 137-144.
Periñán Pascual, C. y Arcas Túnez, F. (2010a), “Ontological commitments in FunGramKB”, Procesamiento del
Lenguaje Natural, 44: 27-34.
Periñán Pascual, C. y Arcas Túnez, F. (2010b). “The Architecture of FunGramKB”, en Proceedings of the Seventh
International Conference on Language Resources and Evaluation, Malta, ELRA, pp. 2667-2674.
Periñán Pascual, C. y Carrión Varela, M.d.l.L. (2011). “FunGramKB y el conocimiento cultural”, en Anglogermanica
Online 2011, pp. 87-105. Disponible en: http://anglogermanica.uv.es:8080/Journal/Viewer.
aspx?Year=2011&ID=pericarrion.pdf [fecha de consulta: 14/10/2013].
Periñán Pascual, C. y Mairal Usón, R. (2009). “Bringing Role and Reference Grammar to natural language
understanding”, Procesamiento del Lenguaje Natural, 43: 265-273.
RLyLA Vol. 09 (2014), 1-13 | 12
Periñán Pascual, C. y Mairal Usón, R. (2010). “La gramática de COREL: un lenguaje de representación conceptual”,
Onomázein, 21: 11-45.
Periñán Pascual, C. y Mairal Usón, R. (2012). “La dimensión computacional de la Gramática del Papel y la
Referencia: la estructura lógica conceptual y su aplicación en el procesamiento del lenguaje natural”, en
R. Mairal Usón, L. Guerrero y C. González (eds.). El funcionalismo en la teoría lingüística. La Gramática del
Papel y la Referencia. Introducción, avances y aplicaciones. Madrid: Akal, pp. 333-348.
Van Valin, R.D. Jr y Mairal Usón, R. (en prensa). “Interfacing the Lexicon and an Ontology in a Linking Algorithm”,
en M. Ángeles Gómez, F. Ruiz de Mendoza y F. Gonzálvez-García (eds.). Form and Function in Language:
Functional, Cognitive and Applied Perspectives. Essays in Honour of Christopher S. Butler. Amsterdam:
John Benjamins.
RLyLA Vol. 09 (2014), 1-13 | 13
Vol. 9 año 2014, 14-22
EISSN 1886-6298
Detection and correction of linguistic errors: results according to
linguistic preferences and uses
Jordi Cicres
Sílvia Llach
M. Dolors de Ribot
Universitat de Girona
Abstract: This paper examines the ability of 256 university students to detect and correct errors related to different linguistic
aspects of Catalan, by means of a test including correct and incorrect sentences. The sentences included aspects of Catalan
prescriptive grammar, errors caused by interference from Spanish, and discursive aspects (adequacy, coherence and cohesion).
In addition to that, the ability of the students has also been linked to data from their language use and linguistic experience in the
following areas: language spoken in the family, language spoken with friends, and formal and informal languages of preference.
The results show that students whose preferred language is Spanish or either Spanish or Catalan are less competent than those
using Catalan on a regular basis. Taking into account the type of error, students who use preferably Spanish or both languages
show significantly worse results in the error detection task (in sentences containing grammar and interference errors) than those
who prefer Catalan. In contrast, no significant differences are shown regarding discourse errors when the students’ linguistic use
is considered.
Keywords: error detection, error correction, Catalan norm, linguistic interference, linguistic use.
1. introduction
The difference between novice and expert writers is a topic that has been studied since the 1970s (Emig, 1977;
Flower and Hayes, 1980; Bereiter and Scardamalia, 1987; among others). It is held that expert writers use a series
of complex cognitive strategies that, apart from grammar skills, also includes aspects related to Psycholinguistics,
Sociology, Pragmatics or Sociolinguistics.
The writing process of expert writers is complex, and it consists of three basic stages: planning, translating
(textualization) and reviewing. Each one of these stages involves the knowledge of multiple skills. Text reviewing
with regard to both the form (superficial revision) and the content (accurate revision) plays a central role in the
writing process (Ferris, 2006; Goldstein, 2006; Sachs and Polio, 2007). In fact, there are many studies that have
focused on various aspects linked to text reviewing: from the process of reviewing in general (García and AriasGundín, 2004; Pérez, 2001; Conrad and Goldstein, 1999; Chanquoy, 1997; Cameron et al., 1997; Camps, 1992) to
specific grammatical features (Largy, Chanquoy and Dédéyan, 2004; Largy, Dédéyan and Hupert, 2004).
But it is obvious that, in order to be expert writers, it is essential for students to be able to review their own
texts critically with regard to the content and the text structure as well as the use of language. This means that
the reviewing process will involve the knowledge of a series of skills. On the one hand, an accurate reviewing will
imply the control of the objectives set, the development of the ideas, and text adequacy, coherence and cohesion,
among others. So the correct use of cohesive elements (of text grammar) will play a significant role. On the other
hand, one of the main problems with respect to the superficial reviewing has to do with the ability to detect
grammatical errors (relative to the prescriptive grammar of the language examined or loan translations borrowed
from other languages) and orthographic errors.
| 14
Jordi Cicres, Sílvia Llach and M. Dolors de Ribot
Detection and correction of linguistic errors: results according to linguistic preferences and uses
From a university perspective, precisely one of the problems observed in Catalan students, and one that
causes more concern, is the lack of knowledge of the language rules. In particular, rules related to aspects of
Catalan itself and aspects derived from the influence of Spanish, or the weak knowledge of textual elements.
Despite all these shortages, Catalan students get results similar to the rest of Spanish students with regard to their
linguistic skills, according to the results from the comparative studies on linguistic skills. Those results are obtained
when the Catalan in Catalonia is compared to the Spanish in the rest of Spain or when what is exclusively analyzed
is the competence in Spanish in both places. Arnau (2004) reviews different studies along these lines (even though
he does not specifically measure the detection and correction of errors).
Firstly, the reading comprehension tests from the PISA report (OCDE, 2001) focus on the access and recovery
of the information, the integration and interpretation of the information, and the reflection and assessment. The
results in Catalonia are always slightly better than the Spanish average, and they are close to the average of the
group of countries studied. Those results are maintained in the reports published on the PISA tests in posterior
tests (MEC, 2007; MEC, 2010).
Secondly, the reports from the Instituto Nacional de Calidad y Evaluación (INCE, 1999; INCE, 2000) analyze the
knowledge of Spanish in Catalonia and in the rest of Spain. They take four dimensions into account: oral and reading
comprehension, writing, knowledge of the use of language and Spanish literature. The general results indicate that
the knowledge of Spanish in Catalonia is comparable to the average in Spain. Finally, the comparative study on
the linguistic skills of students in Catalonia (SEDEC, 1998) compares the results obtained on the knowledge of
Catalan and Spanish. It also provides information on the native language of the students (Catalan or Spanish). This
study focuses on 14 different dimensions: oral comprehension, morphosyntax, spelling, reading comprehension,
writing (adequacy, coherence, cohesion and linguistic correction), oral text (linguistic correction and organization),
phonetics and reading (mechanics, intonation and speed). In the assessment of Catalan, the results are always
higher for Catalan-speaking students (with statistical significance) or similar to Spanish-speaking students. In the
assessment of Spanish, on the other hand, the results are much more homogeneous in both groups. From the 14
dimensions analyzed, there are four with significant differences (in three of which Catalan-speaking students get a
higher score and only in one, oral expression, Spanish-speaking students obtain better results).
These results can be explained in the framework of the Interdependence Hypothesis (Cummins, 1979).
According to this hypothesis, students are capable of transferring some linguistic knowledge learned in one
language to another language. For example, the ability to create coherent, cohesive and adequate texts, or the
ability to make metalinguistic judgements, to explain, narrate, etc.
In order to tackle this issue, this article focuses only on the following aspects of reviewing: the ability to
successfully detect and correct errors. We present empirical results on this skill from 256 Primary Education
undergraduate students (future primary school teachers) in their first year at the University of Girona (during the
2009-2010 academic year). The goal is to analyze this ability and connect it to various data indicating the students’
preferences and linguistic uses.
The test provided consisted of one error detection and correction exercise. Following van der Pool’s (1996)
criteria, who classifies the assessment tools of the reviewing process into online tools (direct observation, interview
while the reviewing is being carried out, thinking out loud or specific detection and correction tests) and offline tools
(questionnaires, interview after the reviewing task or a posteriori evaluations by the teachers of texts produced by
the learners), we can say that our test belongs to the online category, similarly to studies like Cameron et al. (1997),
Hacker et al. (1994), McCutchen et al. (1997) and Wallace et al.(1996), where students have to detect, mark and
correct textual errors.
More specifically, the task designed for this research was based on James (1998) analysis, who divides the
learner’s linguistic errors into three categories: intralinguistic errors (caused by difficulties of the language itself),
interlinguistic errors (due to the influence of other languages) and errors related to discourse properties (including
both discourse errors and pragmatic ones). Thus, our test takes these three categories into account and deals
with the following types of error: a) errors relative to the prescriptive grammar and orthography specific to Catalan
(intralinguistic errors); b) errors related to the linguistic interference of Spanish (interlinguistic errors); and c) errors
connected to discourse adequacy, coherence and cohesion (discourse errors).
The results obtained from the participants are connected to their linguistic uses in the following environments:
language spoken in the family, language spoken with their friends, preferred formal language and preferred informal
language. Hence these data will allow us to know which groups of students have more difficulties detecting and
correcting each type of error.
RLyLA Vol. 09 (2014), 14-22 | 15
2. GOALS AND HYPOTHESES
The present study focuses on the empirical analysis of the ability to detect and correct errors related to the
prescriptive grammar, to the linguistic interference of Spanish and to discourse elements (of coherence, cohesion
and adequacy) and it has the following goals:
a. The analysis of how the students’ ability to detect and correct errors is connected with their everyday
use of Catalan or Spanish. This is done in the following contexts: language normally used in the family;
language normally used with friends; preferred language in formal situations; and preferred language in
informal situations;
b. The finding and analysis of the types of error that offer more difficulties to each group of students.
In order to formulate the hypotheses, we have considered the results from different evaluation studies (such
as SEDEC, 1998; INCE, 1999; INCE, 2000; OCDE, 2001; MEC, 2007; MEC, 2010). So the hypotheses that are
assessed here are the following:
a. Overall, students who prefer and use Catalan more often will obtain better results in detecting and
correcting errors (presented in the Catalan language).
b. Students who have Spanish as their preferred language will show a lower ability to detect errors than
students who prefer Catalan due to linguistic interference.
c. There will be no statistically significant differences in the effectiveness to detect and correct errors related
to Catalan prescriptive grammar, since all the students have had the same formal education in prescriptive
grammar during their obligatory education.
d. There will be no statistically significant differences in the detection and correction of errors related to text
coherence, cohesion and adequacy, since these are more transferable aspects from one language to
another than orthographic or grammatical features.
3. METHODOLOGY
3.1. Participants
The participants in this study were 256 Primary Education first year undergraduate students. The test was
answered at the end of the academic year. With regard to their gender, 64% were female and 36% male. The
access to university was mainly through a high school diploma and a university entrance examination (70.5%),
followed by a vocational qualification (23.4%), examinations for students over 25 years of age (2.8%), and others
(from other university degrees or other types of access, 3.3%). The native language of the participants was Catalan
(65%), Spanish (25.7%) or both languages indistinctively (9.3%). However, since all participants lived and attended
Primary and Secondary schools in Catalonia in their childhood and adolescence, they were supposedly proficient
in both Catalan and Spanish.
3.2. Test design
The test designed consists of 60 sentences. The first 20 sentences are related to the interference of Spanish
grammar (interlinguistic errors). The topics selected correspond to those aspects that are more problematic, as
identified by Amadeo and Solé (1996) and Mestres et al. (2007), and frequent in the writings of university students,
according to the researchers’ preliminary observations. Sentences 21-40, on the other hand, deal with grammatical
and orthographic issues specific to Catalan (intralinguistic errors). The choice of topics was based on the test
of Nivell de suficiència of the Direcció General de Política Lingüística of the Generalitat de Catalunya, which is
equivalent to the C1 level of the Common European Framework of Reference for Languages (CEFR). Finally, the
last sentences (41-60) have to do with discourse matters (adequacy, coherence and cohesion).
The test started with this statement: “In the following sentences, there can be various problems of language
(adequacy, coherence, cohesion, grammar, prescriptive grammar...). Say if the sentence is correct or has problems.
If you think that a sentence has problems, write below the version that you consider better or correct.” In the
appendix we provide the 60 sentences with the particular aspect that is being assessed. The sentences marked
with an asterisk are incorrect or have problems. In four questions, more than 50% of the students gave new
correct versions of already correct sentences (on average, 64% of the students with Catalan L1 and 61% of the
students with Spanish L1). In these cases, we decided to cancel the sentence and not to take them into account
RLyLA Vol. 09 (2014), 14-22 | 16
in the statistical analysis, since the formal context in which the test was handed out could have caused the false
impression that some of the sentences contained hidden difficulties.
The reliability of the test has been statistically validated through the split-half method (Spearman-Brown
coefficient = 0.791; Guttman Split-Half coefficient = 0.787).
In order to obtain the variables referring to the linguistic preferences, the participants were asked to provide
information (through a written form with closed questions) about the language normally used in the family and with
friends, and the preferred language in formal and informal situations. Table 1 shows a summary of the participants’
characteristics according to their linguistic preferences. As can be observed, there is a clear preference for Catalan
in all the environments. The percentage of students who claimed to use other languages (2.5% of the total) was
discarded from the statistical analysis since the sample was not representative.
Table 1. Percentage of participants according to their preferences and linguistic uses.
Language normally used in the family
Catalan (%)
Spanish (%)
Catalan and Spanish
interchangeably (%)
65
25.7
9.3
Language normally used with friends
67.1
12.2
20.7
Preferred language in formal situations
86.1
5.5
7.6
Preferred language in informal situations
70.5
16
13.5
3.3. Statistical analysis
For the statistical treatment of the data, we used contingency tables (with the specification of the corrected
typified residues) and Pearson’s chi-square statistic. Once we analyzed whether or not there was a significant
relationship between the ability to detect and correct the errors and the factors studied, the next important step
was to measure what the direction of that relationship was by comparing the observed cases and the expected
cases. The expected cases are the result that would be obtained in the sample analyzed if the two variables
(acceptable detection and correction of errors, on the one hand, and language use and preferred language in the
various environments studied, on the other) were independent of each other, that is, if there was no relationship
between the presence of errors and the factor analyzed. The fact that there is a significant relationship between the
two variables causes the observed cases of detection of errors registered not to coincide with the expected ones,
precisely because the factor determines that ability. In order to be able to carry out the comparison between cases
expected and cases observed, the adjusted residual values (ARV) were utilized. This statistical value follows a
normal distribution with zero average and standard deviation 1. Therefore, assuming a confidence interval of 95%,
we can say that residual values higher than 1.96 or lower than -1.96 show the situations with the crossing of two
variables containing more or fewer cases, respectively, than the ones expected if the variables were independent.
4. RESULTS
Firstly, table 2 shows the ARV results, which indicate the ability to detect and correct the errors successfully
according to the language used in the various situations examined and the participants’ preferred language in
the study. The p-value of the chi-square is highly significant in all cases. The results demonstrate that students
who claim to use Spanish more than Catalan (or, to a lesser degree, both languages together) have a lower
ability to detect and correct errors than students who only use Catalan (since they show significant negative ARV
values, while students who claim to use Catalan more often show significant positive values). This aspect needs
to be considered since it shows a different profile in those participants who use Spanish on a regular basis. We
can see neither a positive transfer in terms of Odlin (1989) nor facilitation in terms of Ellis (1994) and Ellis and
Barkhuizen (2005), which would involve an improvement in the learning of the two languages due to their structural
similarities, but the performance of Catalan in the students who use Catalan and Spanish or Spanish as their
preferred language is less satisfactory than in those students who mainly use Catalan.
RLyLA Vol. 09 (2014), 14-22 | 17
Table 2. Ability to detect and correct the error successfully according to the preferred language and use. Adjusted residual values
and Pearson’s chi-square. Asterisks show significant ARV values.
Adjusted residual values (ARV)
Catalan
Spanish
Catalan and Spanish
interchangeably
sign. Pearson’s chi-square
language in the family
5.6*
-6.6*
0.7
p<0.001
language with acquaintances
7.2*
-6.8*
-2.9*
p<0.001
preferred formal language
preferred informal language
6.4*
7.9*
-6.9*
-5.7*
-2.2*
-4.5*
p<0.001
p<0.001
Secondly, the analysis of the ability to detect errors according to their typology shows that students who claim
to use Spanish more often than Catalan (or the two languages interchangeably) are less able to identify linguistic
interference errors and errors specific to Catalan grammar than students who prefer the use of only Catalan. With
regard to the detection of errors related to discourse aspects, the groups do not show any differences according
to their preferred language.
Table 3. Ability to detect and correct errors successfully according to the type of error and the preferred language of use in
several situations. CTR and indication of the degree of significance of p.
Linguistic
interference
Specific aspects of the Catalan grammar
Discourse aspects
language
in the
family
Catalan: 6.5
Spanish: -7.8
Both: no sign.
Pearson’s chi-square***
Catalan: 2.7
Spanish: -3.1
Both: no sign.
Pearson’s chi-square**
Pearson’s chi-square no sign.
language
with
friends
Catalan: 7.8
Spanish: -6.7
Both: -3.7
Catalan: 4.8***
Spanish: -4.2***
Both: -2.2***
preferred
formal
language
Catalan: 6.7
Spanish: -6.6
Both: -2.8
Catalan: 4,6
Spanish: -3,7
Both: -2,7
Catalan: no sign.
Spanish: -2
Both: no sign.
Pearson’s chi-square*
preferred
informal
language
Catalan: 8.7
Spanish: -6.4
Both: -4.7
Catalan: 4.6
Spanish: -2.6
Both: -3.4
***p<0.001;**p<0.01;*p<0.05
5. DISCUSSION AND CONCLUSIONS
The first goal set in this study was to analyze the ability of different groups of students to detect and correct
linguistic errors successfully according to their preferences in the use of Catalan or Spanish and the language
normally used in different environments. The observation of the data presented here allows us to state that, in
absolute terms, CTR values higher than 1.96 or lower than -1.96 show significant differences. These data agree
with the findings provided in the comparative study of the knowledge of Catalan and Spanish sponsored by the
Servei d’Ensenyament del Català - SEDEC (1998) (cited in Arnau, 2004), where the knowledge of Catalan and
Spanish among 6th-grade schoolchildren in Catalonia is evaluated (this study has the following variables into
account: language in the family, socio-professional position of the family and the type of school, i.e., state, private
or state-subsidised school). The SEDEC study shows that competence in Catalan (which tests oral comprehension,
morphosyntax, orthography, phonetics, reading mechanics, intonation and reading rate) is significantly better
in Catalan-speaking students than in Spanish speakers in 9 out of 14 tests. In addition to this, it also clearly
demonstrates that the pupils’ family language does not cause any setback in the knowledge of Spanish. We can
see this in the fact that there are no significant differences in most of the tests, and that only in one of the 14 tests
Spanish-speaking students obtain better results than Catalan-speaking students (against three in which Catalanspeaking students get a better score). Hence, the results obtained in our study are in the same line as the results
shown by the SEDEC study. Therefore, hypothesis a) is confirmed.
RLyLA Vol. 09 (2014), 14-22 | 18
As far as the second goal is concerned (i.e., the analysis of which typology of errors shows more difficulties
for each group of students according to their linguistic uses), it can be observed that students that prefer Spanish
get worse results with regard to the detection of errors caused by interference processes (in a high degree) and
errors related to Catalan prescriptive grammar (to a lesser degree, although it is also statistically significant). These
data allow us to validate hypothesis b), but not hypothesis c), since the worst results are maintained in the group
preferring Spanish in the various situations examined.
Related to the latter hypothesis, the results obtained are not those expected as the aspects on normative
grammar that are assessed are specific to the written language and to a formal register (therefore, quite far from
colloquial oral uses). Thus, the use of Catalan in the family should not be an advantage. On the other hand, no
significant differences are seen on the detection of problems relating to the discourse, so the results validate
hypothesis d).
It is interesting to emphasize that the results show a lack of ability to detect orthographic and grammatical
errors, interference errors and discourse errors in all groups, although mainly in the group where Spanish is highly
preferred. This inability prevents the students from getting the skills in writing that are characteristic of expert
writers, among other things because they neglect one of the three fundamental stages in the writing process, i.e.,
correction. Students who prefer Spanish get results that are overall worse than students who prefer Catalan, so
they will need strategies that allow them to improve their communicative competence.
Finally, it has to be stressed that, according to the SEDEC (1998) study, Catalan students (both Catalan speakers
and Spanish speakers) obtain general results on linguistic competence in Spanish similar to those observed in
Spanish monolingual autonomous communities, but, in the former case, students also gain linguistic competence
in Catalan (clearly higher in Catalan speakers). The SEDEC data also indicate that in practically all the tests, both
in Catalan and in Spanish, Catalan speakers get better results than Spanish speakers.
The analysis of the students’ linguistic skills is a topic that needs to be investigated in more detail. This study
contributes in this direction by showing that the ability to detect errors in the Catalan language is higher in Catalan
speaking students even when all the students received the same education. Now a matter for further research
would be to carry out the test the other way around so we could observe the ability of Catalan speakers and
Spanish speakers to detect errors in Spanish.
If the results followed the same direction as our study has pointed out, it would seem that, despite having a
good bilingual or multilingual education, the proficiency of the preferred language in the family and with friends
would be higher than in other languages. In this case, it would be necessary to find strategies to correct these
inequalities in the knowledge of one language so the speakers who do not have it as their preferred one could
overcome this disadvantage (which, as we have seen, it is maintained at university) without affecting the knowledge
of other languages. This would help us to get an overall view of the situation, and it would also mean a step closer
towards understanding how the different skills are transferred from one language to another and how one language
interferes with another.
REFERENCES
Amadeo, I. and Solé, J. (1996). Curs pràctic de redacció. Barcelona: Ed. Columna.
Arnau, J. (2004). “Sobre les competències en català i castellà dels escolars de Catalunya: una resposta a la
polèmica sobre el decret d’hores de castellà”, Llengua, Societat i Comunicació, 1: 1-7.
Bereiter, C. and Scardamalia, M. (1987). The psychology of written composition. Hillsdale (N.J.): Erlbaum.
Cameron, C., Edmunds, G., Wigmore, B. Hunt, A. and Linton, M. (1997). “Children’s revision of textual flaws”,
International Journal of Behavioral Development, 20/4: 667-680. doi:10.1080/016502597385117
Camps, A. (1992). “Algunas observaciones sobre la capacidad de revisión de los adolescentes”, Infancia y
Aprendizaje, 58: 65-81. doi:10.1080/02103702.1992.10822333
Chanquoy, L. (1997). “Thinking skills and composing: Examples of text revision”, in J. H. M. Hamers and M.
Overtoom (ed.) Inventory of European Programmes for teaching thinking, Utrecht: Sardes, pp. 179-185.
Conrad, S. M. and Goldstein, L. M. (1999). “ESL student revision after teacher-written comments: text, contexts
and individuals”, Journal of second language writing, 8/2: 147-179. doi:10.1016/S1060-3743(99)80126-X
Cummins, J. (1979). “Cognitive/academic language proficiency, linguistic interdependence, the optimum age
question and some other matters”, Working Papers on Bilingualism, 19: 121-129.
Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford Applied Linguistics.
Ellis, R. and Barkhuizen, G. (2005). Analysing Learner Language. Oxford: Oxford University Press.
RLyLA Vol. 09 (2014), 14-22 | 19
Emig, J. (1977). “Writing as a mode of learning”, College Composition and Communication, 28: 122-128.
doi:10.2307/356095
Ferris, D. (2006). “Does error feedback help student writers? New evidence on the short- and long-term effects of
written error correction”, in K. Hyland and F. Hyland (ed.) Feedback in second language writing: Contexts
and issues, Cambridge: Cambridge University Press, pp. 81-104. doi:10.1017/CBO9781139524742.007
Flower, L. and Hayes, J. R. (1980). “The dynamics of composing: Making plans and juggling constraints”, in L. W.
Gregg and E. R. Steinberg (ed.), Cognitive processes in writing, Hillsdale: N.J. Erlbaum, pp. 31-50.
García, J. N. and O. Arias-Gundín (2004). “Intervención en estrategias de revisión del mensaje escrito”, Psicothema,
16/2: 194-202.
Goldstein, L. (2006). “Feedback and revision in second language writing: Contextual, teacher, and student
variables”, in K. Hyland and F. Hyland (ed.), Feedback in second language writing: Contexts and issues,
Cambridge: Cambridge University Press, pp. 185-205. doi:10.1017/CBO9781139524742.012
Hacker, D., Plumb, C., Butterfield, E., Quathamer, D. and Heineken, E. (1994). “Text revision: Detection and
correction of errors”, Journal of Educational Psychology, 86/1: 65-78. doi:10.1037/0022-0663.86.1.65
Instituto Nacional de Calidad y Evaluación - INCE (1999). Evaluación de la Educación Primaria. Madrid: Ministerio
de Educación, Cultura y Deporte.
Instituto Nacional de Calidad y Evaluación - INCE (2000). Evaluación de la Educación Secundaria. Madrid:
Ministerio de Educación, Cultura y Deporte
Largy, P., Chanquoy, L. and Dédéyan, A. (2004). “Orthographic revision: the case of subject-verb agreement in
french”, in L. Allal, L. Chanquoy and P. Largy (ed.) Revision cognitive and instructional processes, New York:
Kluwer Academic Publishers, pp. 39-62. doi:10.1007/978-94-007-1048-1_4
Largy,P., Dédéyan, A. and Hupet, M. (2004). “Ortographic revision: a developmental study of how revisers
check verbal agreements in written texts”, British Journal of Educational Psychology, 74: 533-550.
doi:10.1348/0007099042376382
McCutchen, D., Francis, M. and Kerr, S. (1997). “Revising for meaning effects of knowledge and strategy”, Journal
of educational psychology, 89/4: 667-676. doi:10.1037/0022-0663.89.4.667
MEC (2010). PISA 2009. Programa para la evaluación internacional de los alumnos. OCDE. Informe español.
Madrid: Instituto de Evaluación. Ministerio de Educación. doi:10.1371/journal.pone.0011612
MEC (2007). PISA 2006. Programa para la evaluación internacional de los alumnos. OCDE. Informe español.
Madrid: Instituto de Evaluación. Ministerio de Educación.
Mestres, J. M., Costa, J., Oliva, M. and Fité, R. (2007). Manual d’estil. La redacció i l’edició de textos. 3rd edition.
Vic: Eumo.
Organisation for Economic Cooperation and Development - OCDE (2001). “Regard sur l’education. Les
indicateurs de l’OCDE”, in Consell Superior d’Avaluació del Sistema Educatiu (2002), Sistema d’indicadors
d’Ensenyament a Catalunya. Barcelona: Generalitat de Catalunya, Departament d’Ensenyament.
Odlin, T. (1989). Language transfer. Cambridge: Cambridge University Press. doi:10.1017/CBO9781139524537
Pérez, S. A. (2001). “Revising During Writing in a Second Grade Classroom”, Educational Research Quarterly,
25/1: 27-32.
Sachs, R. and Polio, C. (2007). “Learners’ uses of two types of written feedback on a L2 writing revision task”,
Studies in Second Language Acquisition, 29: 67–100. doi:10.1017/S0272263107070039
Servei d’Ensenyament del Català – SEDEC (1998). Competència lingüística en català i castellà de l’alumnat de
sisè de Primària, curs 1997-1998. Barcelona: Generalitat de Catalunya, Departament d’Ensenyament.
Unpublished.
Van der Pool, E. (1996). “A text-analytical study of conceptual writing processes and their development”, in G.
Rijlaarsdam, H. van der Bergh and M. Couzijn (ed.) Theories, Models and Methodology in Writing Research,
Amsterdam: Amsterdam University Press, pp. 456-472.
Wallace, D., Hayes, J., Hatch, J., Miller, W., Moser, G. and Silk, C. (1996). “Better revision in eight minutes?
Prompting first-year college writers to revise globally”, Journal of Educational Psychology, 88: 682-688.
doi:10.1037/0022-0663.88.4.682
APPENDIX
Sentences from the test used in this study and a brief explanation of the errors produced. The asterisks indicate
the sentences that are incorrect and/or containing discourse problems. Sentences (1 to 20) in the test contain
interlinguistic errors derived from Spanish interference; sentences (21 to 40) have errors related to intralinguistic
aspects, that is, derived from difficulties inherent to the Catalan language; and finally, sentences (41 to 60) contain
errors related to discourse aspects (coherence, cohesion, adequacy). Cancelled sentences are not included.
RLyLA Vol. 09 (2014), 14-22 | 20
1. Té tot el que vol. Correct use of the neuter article.
2. No tinc massa sort.* Use of massa (‘too’) in negative sentences.
3. Han trobat un quadre excepcional, però desconeixen l’autor del mateix.* Incorrect use of mateix ‘self’.
5. L’he vist vàries vegades.* Incorrect use of vàries ‘several’.
6. No podien tenir fills, i van adoptar un.* Incorrect deletion of the partitive pronoun en.
7. Ha marxat a casa, doncs tenia mal de panxa.* Incorrect use of doncs ‘then,’ which is assigned a causative
value.
8. No hi havia gaire gent al carrer. Sentence with the correct conjugation of the verb haver-hi ‘there is,’ with no
deletion of hi ‘there.’
9. Quan aniràs a Barcelona? Aniré demà.* Incorrect deletion of the pronoun hi ‘there.’
10. Van començar les obres al gener, acabant-les al juny.* Incorrect use of the gerund, which is assigned the value
of a subsequent action.
11. Es tracta que ho escriguis tu. Correct use of the conjunction que ‘that’, i.e., it is not preceded by a preposition.
12. El tenen que operar.* Incorrect use of the periphrasis of obligation.
13. Hi ha que reconèixer que això està molt bé.* Incorrect use of the periphrasis of obligation.
14. Vol algo més?* Incorrect use of the Spanish word algo ‘something’.
15. Vaig acompanyar l’Oriol a casa seva. Correct sentence, i.e., the direct object is not preceded by the preposition
a ‘to’.
16. Vull veure al meu fill.* Incorrect use of the direct object, since it is preceded by a preposition.
17. Està segur de que el Barça guanyarà.* Incorrect use of a preposition preceding the conjunction que ‘that’.
18. Va comprar dues cases i va donar aquestes als seus dos fills.* Excessive use of the demonstrative.
19. Al entrar ell, tothom va callar.* Al ‘to the’ is an incorrect form to introduce the infinitive with a time value.
20. No hi havia gaire gent degut a la pluja.* Incorrect use of degut a ‘due to’.
21. Al final he entès que la intel·ligència és una capacitat molt complexa. Correct sentence.
22. Aquest és el llibre del que et vaig parlar.* Incorrect use of the relative pronoun.
24. No em pensava que fossis tant tossut.* Incorrect use of tant ‘so much’ instead of the adverb tan ‘so.’
25. Després de l’escàndol, va ser inevitable que desapareixés del panorama musical.* Incorrect verbal form
desapareixés ‘disappeared’ (subjunctive).
26. No volem el pastís, mengeu’s-el vosaltres.* Incorrect use of the combination of weak pronouns.
27. Envia-lis una carta als socis.* Incorrect use of the weak pronoun.
28. Però en aquí hi ha una part negativa que s’ha de comentar.* Insertion of preposition due to the influence of the
spoken language.
29. Cal apendre dels errors.* Incorrect form of the infinitive.
30. Són el tipus de mestre que mostra més seguretat alhora de fer les classes.* Incorrect use of the adverb alhora
‘at the same time’, instead of a l’hora ‘at the time of’.
31. Per aquest motiu la tutora la castigat sense pati.* Incorrect contraction between a pronoun and the auxiliary
verb.
32. Si haguessis tornat, haguessis vist la posta de sol.* Incorrect combination of verbal forms.
33. Prengueu el que vulgueu, que avui convido jo.* Incorrect imperative verbal form.
34. No he vist enlloc un desordre com el que hi ha en aquesta casa. Correct use of enlloc ‘nowhere’.
35. Va arribar tard a l’oficina per què va perdre l’autobús.* Incorrect use of the interrogative form per què ‘why’ in
a sentence with a causative meaning.
36. Quan va arribar la última cançó es van posar neguitosos.* Incorrect use of the apostrophe.
37. Vam preguntar-los quins dibuixos hi havia en el conte. Correct use of the weak pronoun.
38. Quan acaben l’activitat, se’n porten una llibreta a casa.* Incorrect use of the verb and clitics due to a confusion
with the verb emportar-se ‘take away’.
39. Els nous exercicis pot ser que ajudin a desenvolupar aquesta capacitat. Correct use of pot ser ‘maybe’.
40. El treball consistia en decorar la cartolina amb ceres.* Incorrect use of the preposition preceding the infinitive.
41. En primer lloc, us vull felicitar. D’altra banda, us vull dir que tenim molts projectes nous.* Problem of cohesion
due to the inadequate use of the linking word d’altra banda ‘on the other hand’.
42. La Caputxeta va trobar l’àvia quelcom diferent.* Problem of register due to the use of the archaic form quelcom
‘somewhat’ in a children’s tale.
RLyLA Vol. 09 (2014), 14-22 | 21
43. Durant la conversa et va anomenar tres vegades.* Incorrect use of the verb anomenar ‘to mention’ due to a
confusion with the verb nomenar ‘to appoint’.
44. Us envio el currículum amb l’esperança de guanyar-me les garrofes a la vostra empresa.* Problem of register
due to the use of a colloquial form in a formal text.
45. El pàrquing estava complet i al final vaig deixar el cotxe en doble fila. Adequate use of the participle complet
‘completed’.
46. He decidit que tanmateix també hi aniré.* Inadequate use of the adverb tanmateix ‘anyway.’
47. Diu que em vol esperar, tot i que vindré aviat.* Problem of coherence.
48. No m’agrada gens que hi hagi pistatxo a l’amanida. Correct use of pistatxo ‘pistachio nut’.
49. El senador va fer acte de presència al sopar acompanyat de la seva fulana.* Problem of register due to the use
of a colloquial form in a formal text.
50. Ho ha provat d’arreglar per ser al sopar, però no podrà venir. Correct sentence, with no problems of coherence
or cohesion.
51. Li han operat la seva mà dreta.* Excessive use of the possessive determiner.
52. En Joan és alt però simpàtic.* Problem of cohesion due to the wrong use of the adversative conjunction.
53. La pel·lícula ha estat molt interessant. Use of the past participle of the verb estar ‘to be’ as recommended in
the prescriptive grammar.
54. Molts mestres es basen a realitzar la classe llegint i explicant simplement el que el llibre diu.* Unnecessary and
excessive periphrases.
56. Aquest fet s’ha produït pot ser donat que no hi ha haver una explicació prou clara.* Problem of cohesion due to
the combination of the adverb potser ‘maybe’ (incorrectly written) and the linking word donat que ‘given that’.
57. El primer que haig de dir en aquest informe de pràcticum és que m’ha tocat una passada de classe.* Problem
of register due to the use of a colloquial form in a formal text.
59. He fet una relació de les idees del propi tema amb les aportacions que han fet altres grups.* Adequate use of
the adjective propi ‘same’.
60. Era molt bona persona, és a dir, que havia treballat molt.* Problem of coherence and cohesion: the relation
between the two clauses can only be expressed by means of a positive copulative conjunction like i ‘and’.
RLyLA Vol. 09 (2014), 14-22 | 22
Vol. 9 año 2014, 23-32
EISSN 1886-6298
Abstract: Content and Language Integrated Learning (CLIL) is a widely researched approach to foreign language learning and
teaching. One of the pillars of CLIL is the concept of motivation. Some studies have focused on exploring motivation within
CLIL, however there has not been much discussion about the connection between motivation, or other affective factors, and
each component of foreign language learning. Hence, given two groups of learners with the same hours of EFL instruction,
the main objective of this research is to determine whether there exists any kind of interaction between the number of words
learners know receptively and their motivation towards English as a Foreign Language (EFL). Most students in both groups were
highly motivated. No relationship was identified between the receptive vocabulary knowledge and the general motivation for the
secondary graders but a positive significant relationship was found for the primary CLIL graders. Several reasons will be adduced.
Keywords: Motivation, EFL receptive vocabulary, Content and Language Integrated Learning (CLIL), time of instruction.
1. introduction
In the last two decades in the European context, Content and Language Integrated Learning (CLIL) has become
a central issue in the field of foreign language teaching and learning. To provide an accurate definition of CLIL is
an arduous task (Cenoz, Genesee and Gorter, 2013). In the mid 1990s the European Network of Administrators,
Researchers and Practitioners (EUROCLIC) adopted CLIL as the umbrella term that could cover a variety of modes
of content and language teaching implementation, from those more content-oriented to those more languageoriented. A large body of empirical research has explored foreign or second language outcomes within CLIL
provision. For example, Dalton-Puffer (2007) and Dalton-Puffer and Smith (2007) approach CLIL discourse analysis,
Llinares, Morton and Whittaker (2012) focus on classroom interaction, some publications tackle the issue from the
perspective of the different educational levels in which CLIL takes place (e.g., Ruiz de Zarobe and Jiménez Catalán,
2009; Dalton-Puffer, Nikula and Smit, 2010; Fortanet-Gómez, 2013). On the pedagogical side, handbooks for good
implementation are also found, such as Coyle, Hood and Marsh (2010) and Mehisto, Marsh and Frigols (2008).
The affective component has also received attention in the literature, though to a lesser extent than other aspects.
The research conducted by Gabryś-Barker and Bielska (2013), Anglada and Banegas (2012), Lasagabaster (2011),
or Heras and Lasagabaster (in press), for example, underscores the relevance of motivational and affective factors
within a CLIL context as CLIL intends to provide foreign language learning with meaning and real life. The present
study attempts to contribute to this growing literature on affective factors in CLIL. The main goal of the present
paper is to compare the relationship of the outcomes in a receptive vocabulary test and a motivation questionnaire
of two groups of Spanish EFL learners: a group of CLIL 5th primary education learners and a non-CLIL group of
2nd secondary education learners. At the time of data collection, both groups had received the same number of
hours of instruction in English, what cancels out the effect of time of exposure. We will try to deal with these issues
by means of reporting the preliminary results of an investigation in progress.
In the following section, we will present the state of the art of research conducted on the topics involved
here: motivation in foreign language learning and, specifically in CLIL, and receptive vocabulary studies and their
connection with affective factors.
| 23
2. State of the art
2.1 Motivation towards foreign language
The relationship between motivation and foreign language learning has been extensively investigated. Different
models in the literature have addressed this issue. Gardner and Lambert’s (1972) Socio-Psychological Model
pioneered the field. Recently other models have come onto the scene, such as the Self-Determination Theory
(Deci and Ryan, 1985; Noels, Pelletier, Clément and Vallerand, 2000; Noels 2001), or the L2 Motivational Self
System by Dórnyei (2005, 2009). According to Gardner and Lambert (1972), motivation towards language learning
is the desire to achieve a language by means of effort, want or desire, and also affect or attitude. They refer
to orientations, understood as the ultimate goals or reasons behind learning a foreign language. Two types of
orientations are found: integrative orientation or learners’ willingness to learn the language so as to become part of
the target language community, and instrumental orientation, i.e. learners’ desire to command the foreign language
for external reasons. Later models of motivation have postulated other constructs. Deci and Ryan (1985) in the
Self-Determination Theory referred to extrinsic motivation, a type of motivation based on the external factors that
guide the learning of a foreign language, and intrinsic motivation, which relies on the interest that foreign language
learning awakes in the learner. Caution must be applied when comparing these pairs of terms, since they are
multifaceted notions with not so clear-cut boundaries (Carreira, 2005).
Many authors have found a positive correlation between motivation and foreign language attainment (e.g.,
Schmidt and Watanabe, 2001; Masgoret and Gardner, 2003; Csizér and Dörnyei, 2005; Bernaus and Gardner,
2008; Yu and Watkins, 2008). This type of studies have examined the relationship between types of motivation
and foreign language learning by paying attention to variables such as age or gender. Contradictory results are
obtained. As for age, it is found that intrinsic motivation decreases with schooling time whereas extrinsic motivation
increases (Eccles, Wigfield and Schiefele, 1998; Doiz, Lasagabaster and Sierra, 2013). On the contrary, Lepper,
Sethi, Dialdin and Drake, (1997) identified a decrease in intrinsic motivation but not in the extrinsic one, while
Carreira (2006, 2010) found that both types decrease with age.
2.2 Motivation in CLIL
Motivation is considered one of the pillars of CLIL implementation and frequently it also serves to justify it. It is
rare that primary or secondary education students feel attracted towards the language; yet it is much more likely that
they feel attracted towards a content subject. In CLIL the content subject is the excuse to push students towards
the foreign language without noticing it. As Banegas (2013:94) points out, “foreign language learning becomes
an engaging activity when knowledge of the world is approached through it”. Very often, learners’ actual levels of
motivation under CLIL are taken for granted in the literature on the topic. Very often, publications theorize about
how CLIL and its associated methodologies could motivate students and offer guidelines to work motivational
factors (e.g., Mehisto, Marsh and Frigols, 2008; Coyle, Hood and Marsh, 2010). Action or classroom research
becomes a very useful pool for obtaining first-hand knowledge on the processes involved in the interaction of
language learning and motivation (Anglada and Banegas, 2012; Banegas, 2013).
However, there is little systematic empirical evidence on how motivation and other affective factors actually
interact with CLIL pedagogies. Cenoz, Genesee and Gorter (2013:15), for instance, note this fact and agree with
Bruton (2011) on the idea that “student motivation might be reduced because of loss of self-esteem when students
are required to use a language they do not know, and use of the language might actually diminish if the subject
matter is novel and/or complex resulting in reduced language acquisition”. A point that deserves attention is
the fact that the typical CLIL student profile corresponds to one who, before enrolling in CLIL, is academically
motivated towards the foreign language. There is no other way to find out these issues but by means of empirical
research. The research in the Basque Country by (e.g., Lasagabaster, 2011) contributes to palliate the lack of
empirical findings. Some studies have been conducted in Ireland (Murtagh, 2007) and in Finland (Seikkula-Leino,
2007). Overall, results confirm what assumptions suggest: language learning and motivation benefit from each
other in a CLIL context. However, further research is in need to clarify this relationship.
2.3 Receptive vocabulary knowledge
Foreign language vocabulary knowledge is an increasingly important area in the field of Applied Linguistics
(e.g. Schmitt, 2000; Qian, 2002, López Mezquita, 2005; Nation, 2006). Central to the investigation within foreign
language vocabulary knowledge is the distinction between productive and receptive vocabulary knowledge
types. Receptive vocabulary is understood as a passive skill which involves the perception of a word and the
understanding of its meaning in listening and reading. Productive vocabulary refers to an active skill that covers
word production so as to match the speakers’ intention in writing and speaking (Nation 2001).
Receptive vocabulary studies show that the size or breadth of knowledge, i.e., the number of words that
learners know, depends on proficiency as well as exposure to the foreign language and frequency of vocabulary
input (Fan, 2000; Golberg, Paradis and Crago, 2008). The type of instruction is having some bearing on vocabulary
RLyLA Vol. 09 (2014), 23-32 | 24
results, since CLIL seems to offer repeated exposure to new words as well as contextualisation by means of rich
and meaningful content, as Xanthou (2010) explains.
A well-known instrument to measure receptive vocabulary knowledge is the Vocabulary Levels Test (VLT) by
Nation (1983, 1990), largely validated in research (Schmitt and Meara, 1997; Laufer, 1998; Laufer and Paribakht,
1998; Cobb and Horst 1999; Jiménez and Terrazas 2005-2008; Agustín Llach and Terrazas Gallego, 2012). It is a
word-definition matching format test that measures receptive vocabulary breath based on the subjects’ recognition
of words of graded frequency lists of 2,000, 3,000, 5,000, The Academic Word List and the 10,000 most frequent
words in English. Knowing words in a frequency band implies knowing words in all lower bands. Overall, the results
in VLT studies point to the fact that the L2 proficiency level or the number of hours of L2 exposure affects the
number of words that learners know receptively.
2.4. Motivation and vocabulary learning
As motivation has an effect on language learning, a logical assumption could therefore be made that motivation
is likely to facilitate vocabulary learning. So far, however, there has been little discussion about the connection
of motivation and vocabulary learning (Elley, 1989; Gardner and MacIntyre, 1991; Laufer and Husltijn, 2001; Kim,
2008). For example, Laufer and Hulstijn (2001) theorise on the cognitive and motivational load of vocabulary
tasks. As they call it, the Involvement Load Hypothesis, is further investigated by Kim (2008), who finds a
connection between motivational factors and lexical performance. Other studies conclude that both integrative
and instrumental motivation can help vocabulary learning (e.g. Gardner and MacIntyre, 1991). In a different line,
Tseng and Schmitt (2008) propose a framework that explores vocabulary knowledge and motivation and suggest
that motivated vocabulary learning follows a developmental mode and functions as a cyclic process as learners’
motivation towards vocabulary learning ebbs and flows over a time period.
In understanding the effect of motivation on vocabulary acquisition we should also consider the productivereceptive distinction of vocabulary types explained above. As authors such as Laufer and Paribakht (1998) and
Webb (2008) note, production is a more demanding task than reception, an aspect with implications for learners’
motivation towards learning a foreign language. In this vein, Nation (2001:28) concludes that, differently from
receptive vocabulary, in productive vocabulary, if we want to convey a message, we need to have a sense of
wanting to do it. This feature is not required in receptive vocabulary. Then, each type seems to require a different
level of motivation.
3. Purpose
Given a CLIL and a non-CLIL group of students with the same number of hours of instruction in English as a
Foreign Language (839 hours) we attempt to explore (1) the levels of motivation of both groups of learners, (2) their
EFL receptive vocabulary size, and (3) the connection between motivation and receptive vocabulary size in each
group.
4. Method
4.1. Participants
A group of 183 Spanish-speaking learners of 2nd grade of secondary education (aged around 13-14 years old)
and a group of 55 Spanish-speaking EFL learners of 5th grade of primary education (aged around 10-11 years old)
participated in this study. They all belonged to school centers located in the north of Spain.
At the testing time, both groups had received approximately 839 hours of instruction in English as a Foreign
Language. Each group had been exposed to EFL in the English Language Classroom, but the primary group had
also received extra hours of EFL through a CLIL subject. A non-CLIL 5th primary group receives approximately 524
hours of English through traditional EFL classes. The difference of hours between a non-CLIL group and a CLIL
group in 5th year corresponds to the hours in the foreign language that the CLIL group receives via the content
subject class.
4.2. Data gathering instruments, procedures, and analysis
In the present study, the EFL vocabulary size is measured through version 2 of the 2,000-word frequencyband from the receptive version of the Vocabulary Levels Test (2K VLT) by Schmitt, Schmitt and Clapham (2001)
(Appendix 1). The test consists of ten groups of six words and three definitions per group. The test-takers have
to match each of the target word to its corresponding definition. Correct matching is given one point, so that the
maximum score of the test is 30 points. In order to calculate students´ word estimates, Nation’s (1990:78) formula
RLyLA Vol. 09 (2014), 23-32 | 25
“Vocabulary size = N correct answers multiplied by total N words in dictionary (the relevant word list) divided by N
items in test” is applied. We gathered data in one regular school time session. Testees had 10 minutes to complete
the test. Before they started, we gave them clear instructions in their mother tongue both orally and in written form.
On the other hand, we assess learners’ motivation towards EFL by means of a semantic differential technique
of 7-point bipolar rating scale using the following 7 pairs of bipolar adjectives: ‘necessary/unnecessary’,
‘ugly‘/‘nice‘, ‘attractive‘/‘unattractive‘, ‘pleasant‘/‘unpleasant‘, ‘important‘/‘unimportant‘, ‘useful‘/‘useless‘, and
‘interesting‘/‘boring‘. The pair ‘difficult’ / ‘easy’ is also included as a distractor as it does not measure motivation.
General motivation is tested through the 7 pairs of adjectives. These adjectives are introduced with the Spanish
phrase “Considero que el inglés es...” (“I consider English to be…”). This scale (Appendix 2) is part of a questionnaire
adapted from Gardner’s (1985) Attitude/Motivation Test Battery (A/MTB). Intrinsic motivation is measured
through the pairs ‘ugly’/’nice’, ‘attractive’/’unattractive’, ‘pleasant’/’unpleasant’, and ‘interesting’/’boring’; the
extrinsic motivation is measured through the pairs ‘necessary’/’unnecessary’’, important’/’unimportant’, and
‘useful’/’useless’. The complete questionnaire was completed in 10 minutes. Data from the VLT and the motivation
scale were analyzed through SPSS program version 19.0.
5. Results
RQ1. Levels of motivation of both groups of learners.
We arranged learners’ motivation scores according to a three-level scale: level 1 (marks: 1.0 to 3.0), level 2
(marks: 3.01 to 5.0), and level 3 (marks: 5.01 to 7.0), where 1 is the lowest level of motivation and 7 the highest.
Table 1 shows that most students in both groups were highly motivated: 66.48% of the 2nd secondary non-CLIL
graders and 76.36% of the 5th primary CLIL graders.
Table 1. Motivation levels: frequency and percentage.
2nd secondary education (no-CLIL)
5th primary education (CLIL)
Motivation levels
n
%
n
%
Level 1 (marks 1.0-3.0)
121
66.48%
42
76.36%
58
31,87%
13
23.64%
3
1.65%
0
0
Missing (students who do not provide an answer) 1
0
With respect to the scores in general, intrinsic and extrinsic motivation, the primary education CLIL group
overcomes the secondary education non-CLIL group in all measures, as Table 2 indicates.
Table 2. General, intrinsic and extrinsic motivation: descriptive statistics.
General motivation
Intrinsic motivation
Extrinsic motivation
Grade
Mean
Min.
Max.
S.D.
5.41
1
7
1.01
5.70
3.75
7
0.87
4.74
1
7
1.41
5.42
3.00
7
1.16
6.41
1
7
0.97
6.60
3.67
7
0.70
RQ2. EFL receptive vocabulary size in both groups.
Table 3 presents the mean results of both groups in the Schmitt, Schmitt and Clapham’s (2001) version 2 of
the 2,000-word frequency-band from the receptive version of the Vocabulary Levels Test. The secondary group
overcame the primary group (14.83 vs. 10.58). Furthermore, in the primary education group the minimum number
of words the learners understood in the VLT was 1, whereas that for the secondary group the minimum was 4.
In the latter, no learners produced fewer words than 4. As for the maximum number of words, this was 19 in the
primary group and 26 in the secondary group. According to Nation’s (1990:78) formula “Vocabulary size = N
RLyLA Vol. 09 (2014), 23-32 | 26
correct answers multiplied by total N words in dictionary (the relevant word list) divided by N items in test”, the
mean vocabulary size or mean word estimates for the secondary group was 985 words, whereas that for the
primary group was 705 words. Table 4 displays the results related to these word estimates. It is interesting to note
that standard deviation (SD) is high in both groups, what proves the heterogeneity of the samples, as, for example,
in the 2nd secondary group some learners know 1733 words and others just 266, or in the 5th primary group some
learners reach up to 1267 words and some learners only 67.
Table 3. 2K VLT: descriptive statistics.
Mean
Min.
Max.
SD
14.83
10.58
4
26
4.70
1
19
4.04
Mean
Min.
Max.
SD
985
266
1733
312
705
67
1267
269
Table 4. Word estimates: descriptive statistics.
RQ3. Connection between motivation and receptive vocabulary size in each group.
In exploring the relationship between both variables, for the 5th primary education group, we performed a
Shapiro-Wilk normality test to identify the normality of the data set in the VLT results (w = 0.97, p-value = 0.43).
However, the normality cannot be accepted neither in the case of general motivation (w = 0.95, p-value = 0.03),
intrinsic motivation (w = 0.94, p-value = 0.01), nor in extrinsic motivation (w = 0.63, p-value = 1.621202e-10).
The 2nd secondary education sample is bigger than the primary education sample, then, the Shapiro-Wilk
normality test cannot be run. We performed a Lilliefors (Kolmogorov-Smirnov) normality test instead both in the
VLT (d = 0.07, p-value = 0.00), in general motivation (d = 0.10, p-value = 0.00), in intrinsic motivation (d = 0.14,
p-value = 1.382563e-09) and in extrinsic motivation (d = 0.39, p-value = 2.21979e-82). We can accept normality
in no case.
Hence, we ran a Spearman correlation both in the VLT and motivation in both groups of students. As Table 5
shows, only a positive significant relationship was found between the receptive vocabulary knowledge and the
general motivation of the primary CLIL group (p-value < 0.05). Spearman’s rank correlation coefficient measures the
strength of relationship between these two variables, and according to it, we should point to a positive relationship
but a weak one (rho = 0.27). No correlation was identified for the rest of measures.
Table 5. Correlation motivation and 2K VLT (Spearman rank correlation).
General motivation
Intrinsic motivation
and 2K VLT
Extrinsic motivation
and 2K VLT
rho
p-value
rho
p-value
rho
p-value
0.03
0.67
0.00
0.91
0.06
0.40
0.27
0.04
0.19
0.16
0.22
0.10
6. DISCUSSION
The first research question of the present study addresses the issue of the level of motivation (general, intrinsic
and extrinsic) of the two samples of learners. High levels of motivation and quite similar distribution patterns across
levels of motivation are identified in both groups. However, the 5th primary education group slightly surpasses the
secondary education group in the three measures of motivation. Learners in both groups are more extrinsically
than intrinsically motivated. The biggest difference is in intrinsic motivation in favour of the primary graders. These
results are partially in line with the study by Doiz, Lasagabaster and Sierra (2013), who concluded that younger
learners are more intrinsically motivated than older learners, although they find that older learners are more
instrumentally motivated. Further longitudinal studies are required to explore the evolution of these motivation
levels and to corroborate or refute results by these authors or others, such as Lepper, Sethi, Dialdin, and Drake
(1997), who pointed to a decrease in intrinsic motivation but not in the extrinsic one along time, or Carreira (2010,
2006), who concluded that both types decrease with age. On the other hand, whether the type of instruction is
RLyLA Vol. 09 (2014), 23-32 | 27
playing some role on motivation or not is an issue that could be approached again by means of a longitudinal
study. Whatever the case, the present study shows that CLIL is not affecting negatively the learners’ motivational
level towards foreign language acquisition.
The second research question focuses on the learners’ size of receptive vocabulary knowledge, in other words,
the number of words they know receptively. We have found that, as could be expected, the 2nd secondary nonCLIL group overcomes the 5th primary CLIL group. However, the results of our 5th primary CLIL group are better
than the results achieved by a 6th primary EFL group (word estimates: 663, EFL hours: 629) in a study conducted
by Agustín Llach and Terrazas Gallego (2012) in a similar educational context in the north of Spain. In explaining
this result, we should point to the effect of age (or cognitive level) or the type of instruction since the EFL hours of
instruction are kept the same in both groups. It would be interesting to plan a research design to identify, on the
one hand, the role played by the CLIL approach and, on the other, by the exposure time to EFL. We should also
add that, overall, the scores by our groups fall short of the 1,000 most frequent words in English. An important
number of studies that explore receptive vocabulary size of learners in contexts different from the context of the
present study identify better scores in fewer hours of instruction (e.g., Milton and Meara 1998, Takala, 1984).
The third research question of this research addresses the relationship between motivation and VLT results
in both groups of learners. In general, no conclusive evidence is found. There is no correlation between these
variables in the group of secondary non-CLIL graders, what implies that the level of motivation does not necessarily
affect the VLT score and vice versa. It is interesting to mention here, though, that a significant correlation was found
between both variables in 180 students of our sample as they move up a grade (Fernández Fontecha and Terrazas
Gallego, 2012). Also, Fernández Fontecha (2010) found a positive connection between motivation and productive
vocabulary in a group of 250 2nd secondary graders, to which the sub-sample of the present study belongs. A
possible interpretation of this result would be that productive tests there could be a sense of wanting to express
a meaning. As Nation (2001: 28) points, students need to be highly motivated to produce words because this is a
more demanding task than recognizing words. As regards the CLIL group, general motivation level correlates with
VLT results: the higher the general motivation, the higher the receptive vocabulary size is or vice versa. However,
although positive, the correlation is not big, which implies that other variables could be affecting this result. Then,
again, we should consider the type of vocabulary learning (receptive) measured in this study as one probably less
dependent on affective factors, such as motivation, and more dependent on other variables, such as intelligence
or aptitude.
7. CONCLUSION
The study described here set out to mainly explore the connection between the motivation levels and receptive
vocabulary size of two groups of learners of different ages but with the same EFL exposure time: a CLIL group
of 5th grade primary learners and a non-CLIL group of 2nd grade secondary learners. The findings suggest that,
in general, (1) both groups are highly motivated, (2) similar distribution patterns of motivation levels are found in
both groups, (3) once the effect of the time of exposure to the foreign language is cancelled out, the age – or type
of instruction – reveals itself as a determining factor for vocabulary size, and (4) some correlation is identified
between the mean general motivation and receptive vocabulary size only in the case of CLIL students. In general,
it seems that motivation is linked to a need of communication which is present in productive vocabulary tasks but
is somehow missing in receptive vocabulary tasks or tests. Finally, a number of important limitations need to be
considered. First, the current research was not designed to determine which of the two variables, age or type of
instruction, is playing a bigger role in results on motivation and vocabulary learning. Also, longitudinal studies are
required to identify the evolution of the relationship of motivation and vocabulary learning, as they are dynamic and
fluctuating processes. As correlation between motivation and vocabulary learning depends on the task, the effect
of different productive vocabulary tasks and different receptive vocabulary tasks in vocabulary learning results
could be addressed. These studies could be complemented with qualitative research that offers a more complete
picture of learners’ performance. Despite its caveats and exploratory nature, this research has gone some way
towards enhancing our understanding of the interrelation between a type of vocabulary learning (receptive), and
motivation, and thus, contributes to the scarcity of research on vocabulary learning and affective factors.
ACKNOWLEDGEMENTS
This study is part of the research project “Factores individuales y contextuales en la adquisición y desarrollo de
la competencia léxica en inglés como lengua extranjera” funded by the Spanish Ministerio de Ciencia e innovación
(Grant Ref. Nº: FFI2010-19334/FILO). We are also very thankful to our statistician Montserrat San Martin for
providing us with very valuable statistical assistance and helpful insights into data interpretation. Any remaining
errors are our own.
RLyLA Vol. 09 (2014), 23-32 | 28
REFERENCES
Agustín Llach, M.P. and Terrazas Gallego, M. (2012). “Vocabulary Knowledge Development and Gender Differences
in a Second Language”, ELIA, 12: 45-75.
Anglada, L., and Banegas, D. (2012). Views on Motivation and Autonomy in ELT: Selected Papers from the XXXVII
FAAPI Conference. San Martín de los Andes: APIZALS.
Banegas, D. (2013). “The integration of content and language as a driving force in the EFL lesson”, in E. Ushioda
(ed.), International Perspectives on Motivation: Language Learning and Professional Challenges. Basingstoke:
Palgrave Macmillan. pp. 82-97. doi:10.1057/9781137000873.0010
Bernaus, M. and Gardner, R.C. (2008). “Teacher Motivation Strategies, Student Perceptions, Student Motivation, and
English Achievement”, The Modern Language Journal, 92/3: 387-401. doi:10.1111/j.1540-4781.2008.00753.x
Bruton, A. (2011). “Is CLIL so beneficial or just selective? Re-evaluating some of the research”, System, 39: 523532. doi:10.1016/j.system.2011.08.002
Carreira, J.M. (2005). “New framework of Intrinsic/Extrinsic and Integrative/Instrumental Motivation in Second
Language Acquisition”, The Keiai Journal of International Studies, 16: 39-64.
Carreira, J.M. (2006). “Motivation for learning English as a foreign language in Japanese elementary schools”, JALT
Journal, 28: 135-157
Carreira, J.M. (2010). “Relationship between motivation for learning EFL and intrinsic motivation for learning in general
among Japanese elementary school students”, System, 39: 90-102. doi:10.1016/j.system.2011.01.009
Cenoz, J., Genesee, F. and D. Gorter (2013). “Critical analysis of CLIL: taking stock and looking forward“, Applied
Linguistics, 13: 1-21.
Cobb, T. and Marlise E.H. (1999). “Vocabulary sizes of some city University students”, Journal of the Division of
Language Studies of City University of Hong Kong 1.1.
Coyle, D., Hood, P., and Marsh, D. (2010). CLIL: Content and Language Integrated Learning. Cambridge: CUP.
Csizér, K. and Dörnyei, Z. (2005). “The Internal Structure of Language Learning Motivation and its
relationship with Language Choice and Learning Effort.” Modern Language Journal, 89: 19-36.
doi:10.1111/j.0026-7902.2005.00263.x
Dalton-Puffer C., Nikula, T. and Smit, U. (eds.) (2010). Language Use and Language Learning in CLIL Classrooms.
Amsterdam: Benjamins. doi:10.1075/aals.7
Dalton-Puffer, C. (2007) Discourse in Content and Language Integrated Learning (CLIL) Classrooms. Amsterdam:
John Benjamins. doi:10.1075/lllt.20
Dalton-Puffer, C. and Smit, U. (eds.) (2007). Empirical Perspectives on CLIL Classroom Discourse. Frankfurt: Peter
Lang.
Deci, E.L. and Ryan, R.M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. New York:
Plenum Publishing Co. doi:10.1007/978-1-4899-2271-7
Doiz, A., Lasagabaster, D. and Sierra, J.M. (2013). The impact of individual and contextual variables on motivation.
Paper presented at the Applied Linguistics Perspectives on Content and Language Integrated Learning
(ALP-CLIL), Miraflores de la Sierra, Madrid, June 2013.
Dörnyei, Z. (2005). The Psychology of the Language Learner: Individual Differences in Second Language Acquisition.
Mahwah, NJ.: Lawrence Erlbaum.
Dörnyei, Z. (2009). “The L2 motivational self system”, in Z. Dörnyei and E. Ushioda (eds.). Motivation, Language
Identity and the L2 Self. Bristol: Multilingual Matters. pp. 92-142.
Eccles, J.S., Wigfield, A. and Schiefele, U. (1998). “Motivation to succeed”. W. Damon and N. Eisenberg (eds.)
Handbook of Child Psychology, Vol. 3. New York, NY: John Wiley and Sons. pp. 1017-1095.
Elley, W.B. (1989). “Vocabulary Acquisition from Listening to Stories”, Reading Research Quarterly, 24: 174-187.
doi:10.2307/747863
Fan, M. (2000). “How big is the gap and how to narrow it? An investigation into the active and passive vocabulary
knowledge of L2 learners”, RELC Journal, 31: 105-119. doi:10.1177/003368820003100205
Fernández Fontecha, A. (2010). “Gender and motivation in EFL vocabulary production”, in R.M. Jiménez Catalán
(ed.) Gender Perspectives on Vocabulary in Foreign and Second Languages, London: Palgrave Macmillan.
pp. 93-116.
Fernández Fontecha, A. and Terrazas Gallego, M. (2012). The role of motivation and age in vocabulary knowledge.
VIAL, 9: 39-62.
Fortanet-Gómez, I. (2013). CLIL in Higher Education. Towards a Multilingual Language Policy. Bristol: Multilingual
Matters.
Gabryś-Barker, D. and Bielska, J. (2013). The Affective Dimension in Second Language Acquisition. Bristol:
Multilingual Matters.
RLyLA Vol. 09 (2014), 23-32 | 29
Gardner, R.C. (1985). Social Psychology and Second Language Learning. London: Edward Arnold.
Gardner, R.C. and Lambert, W.E. (1972). Attitudes and Motivation in Second Language Learning. Rowley, MA:
Newbury.
Gardner, R.C. and MacIntyre, P.D. (1991). “An instrumental motivation in language study: who says it isn’t
effective?”, Studies in Second Language Acquisition, 13/1: 57-72. doi:10.1017/S0272263100009724
Golberg, H., Paradis, J., and Crago, M. (2008). “Lexical acquisition over time in minority first language children learning
English as a second language”, Applied Psycholinguistics, 29: 41-65. doi:10.1017/S014271640808003X
Heras, A. and Lasagabaster, D. (in press). “The impact of CLIL on affective factors and vocabulary learning”,
Language Teaching Research. Jiménez Catalán, R.M. and Terrazas, M. (2005-2008). “The receptive vocabulary of English foreign language young
learners”, Journal of English Studies, 5-6: 173-191.
Kim, Y. (2008). “The role of task-induced involvement and learner proficiency in L2 vocabulary acquisition”,
Language Learning, 58/2: 285-325. doi:10.1111/j.1467-9922.2008.00442.x
Lasagabaster, D. (2011). “English achievement and student motivation in CLIL and EFL settings”, Innovation in
Language Learning and Teaching, 5: 3-18. doi:10.1080/17501229.2010.519030
Laufer, B. (1998). “The development of passive and active vocabulary in a second language: same or different?”,
Applied Linguistics 19/2, 255-271. doi:10.1093/applin/19.2.255
Laufer, B. and Hulstijn, J. (2001). “Incidental vocabulary acquisition in a second language: the construct of taskinduced involvement”, Applied Linguistics, 22: 1-26. doi:10.1093/applin/22.1.1
Laufer, B. and Paribakht, T.S. (1998). “The relationship between passive and active vocabularies: Effects of
language learning context”, Language Learning, 48: 365-391. doi:10.1111/0023-8333.00046
Lepper, M., Sethi, S., Dialdin, D. and Drake, M. (1997). “Intrinsic and extrinsic motivation: A developmental
perspective”, in S.S. Luthar, J.A. Burack, D. Cicchetti and J.R. Weisz (eds.). Developmental Psychology:
Perspectives on Adjustment, Risk, and Disorder. New York: Cambridge University Press. pp. 23-50.
Llinares, A., Whittaker, R. and Morton, T. (2012). The Roles of Language in CLIL. Cambridge: CUP.
López-Mezquita, M.T. (2005). La Evaluación de la Competencia Léxica: Tests de Vocabulario. Su Fiabilidad y
Validez. Universidad de Granada: Doctoral Dissertation.
Masgoret, A-M. and Gardner R.C. (2003). “Attitudes, motivation, and second language learning: meta-analyses of
studies by Gardner and associates”, in Z. Dörnyei (ed.) Attitudes, Orientations and Motivations in Language
Learning. Oxford: Blackwell. pp. 167-210.
Mehisto, P., Marsh, D., Frigols, M.J. (2008). Uncovering CLIL: Content and Language Integrated Learning in
Bilingual and Multilingual Education. Oxford: Macmillan.
Milton, J. and Meara, P. (1998). “Are the British really bad at learning foreign languages?”, Language Learning
Journal, 18: 68-76. doi:10.1080/09571739885200291
Murtagh, L. (2007). “Out-of-school use of Irish, motivation to learn the language and proficiency in immersion and
subject-only post-primary programmes”, International Journal of Bilingualism and Bilingual Education, 10/4:
428-453. doi:10.2167/beb453.0
Nation, P. (1983). “Teaching and testing vocabulary”, Guidelines, 5: 12-25.
Nation, P. (1990). Teaching and learning vocabulary. New York: Newbury.
Nation, P. (2001). Learning Vocabulary in Another Language. Cambridge: Cambridge University Press. doi:10.1017/
CBO9781139524759
Nation, P. (2006). “How large a vocabulary is needed for reading and listening?”, The Canadian Modern Language
Review/La revue canadienne des langues vivantes, 63/1: 59-81. doi:10.3138/cmlr.63.1.59
Noels, K.A. (2001). “Learning Spanish as a Second Language: Learners’ orientations and perceptions of their
teachers’ communication style“, Language Learning, 51: 107-144. doi:10.1111/0023-8333.00149
Noels, K.A., Pelletier, L.G., Clément, R., and Vallerand, R.J. (2000). “Why are you learning a second
language? Motivational orientations and Self-Determination Theory“, Language Learning, 50: 57-85.
doi:10.1111/0023-8333.00111
Qian, D. (2002). “Investigating the relationship between vocabulary knowledge and academic reading performance:
An assessment perspective”, Language Learning, 52/3: 513-36. doi:10.1111/1467-9922.00193
Ruiz de Zarobe, Y. and Jiménez Catalán, R.M. (eds.) (2010). Content and Language Integrated Learning: Evidence
from Research in Europe. Bristol: Multilingual Matters.
Schmidt, R. and Watanabe, Y. (2001). “Motivation, strategy use, and pedagogical preferences in foreign language
learning”, in Z. Dörnyei and R. Schmidt (eds.) Motivation and Second Language Acquisition (Technical Report
#23). Honolulu: University of Hawai‘i, Second Language Teaching and Curriculum Center. pp. 313-359.
RLyLA Vol. 09 (2014), 23-32 | 30
Schmitt, N. (2000). “Vocabulary acquisition”, in N. Schmitt (ed.) Vocabulary in Language Teaching. Cambridge:
Cambridge Language Education. pp. 116-141.
Schmitt, N. and Meara, P. (1997). “Researching vocabulary through a word knowledge framework: word associations
and verbal suffixes”, Studies in Second Language Acquisition, 19: 17-35. doi:10.1017/S0272263197001022
Schmitt, N., Schmitt, D. and Clapham, C. (2001). “Developing and exploring the behaviour of two new versions of
the Vocabulary Levels Test”, Language Testing, 18: 55-88. doi:10.1177/026553220101800103
Seikkula-Leino, J. (2007). “CLIL learning: achievement levels and affective factors”, Language and Education, 21/4:
328-341. doi:10.2167/le635.0
Tseng, W-T. and Schmitt, N. (2008). “Toward a Model of Motivated Vocabulary Learning: A Structural Equation
Modeling Approach”, Language Learning, 58/2: 357-400. doi:10.1111/j.1467-9922.2008.00444.x
Webb, S. (2008). “Receptive and productive vocabulary sizes of L2 learners”, Studies in Second Language
Acquisition, 30/1: 79-95. doi:10.1017/S0272263108080042
Xanthou, M. (2010). “Current trends in L2 vocabulary learning and instruction. Is CLIL the right approach?”,
Advances in Research on Language Acquisition and Teaching: Selected Papers, Thessaloniki, Greece:
Greek Applied Linguistics Association (GALA). pp. 459-471.
Yu, B. and Watkins, D.A. (2008). “Motivational and cultural correlates of second language acquisition: an investigation
of international students in the universities of the people’s Republic of China”, Australian Review of Applied
Linguistics, 31/2: 17.1-17.22.
Appendix 1.
2,000-word frequency-band from the receptive version of the Vocabulary Levels Test (2K VLT) (Schmitt, Schmitt
and Clapham 2001).
EXAMPLE
ANSWER
1 business
1 business
2 clock
_____ part of a house
3 horse
_____ animal with 4 legs
5 shoe
_____ something used for writing
è
6 wall
2 clock
__6__ part of a house
3 horse
__3__ animal with 4 legs
5 shoe
__4__ something used for writing
6 wall
1 coffee
1 adopt
2 disease
_____ money for work
2 climb
_____ go up
3 justice
_____ a piece of clothing
3 examine
_____ look at closely
4 skirt
_____ using the law in the right way
4 pour
_____ be on every side
5 stage
5 satisfy
6 wage
6 surround
1 choice
1 bake
2 crop
_____ heat
2 connect _____ join together
3 flesh
_____ meat
3 inquire _____ walk without purpose
4 salary
_____ money paid regularly for doing a job
4 limit _____ keep within a certain size
5 secret
5 recognize
6 temperature
6 wander
RLyLA Vol. 09 (2014), 23-32 | 31
1 cap
1 burst
2 education
_____ teaching and learning
2 concern
_____ break open
3 journey
_____ numbers to measure with
3 deliver
_____ make better
4 parent
_____ going to a far place
4 fold _____ take something to someone
5 scale
5 improve
6 trick
6 urge
1 attack
1 original
2 charm
_____ gold and silver
2 private
_____ first
3 lack
_____ pleasing quality
3 royal
_____ not public
4 pen
_____ not having something
4 slow
_____ all added together
5 shadow
5 sorry
6 treasure
6 total
1 cream
1 ancient
2 factory
2 curious
_____ part of milk
_____ not easy
3 nail
_____ a lot of money
3 difficult
_____ very old
4 pupil
_____ person who is studying
4 entire
_____ related to God
5 sacrifice
5 holy
6 wealth
6 social
Appendix 2.
Likert scale adapted from the Attitude/Motivation Test Battery (A/MTB) Gardner’s (1985).
Necesario
Innecesario
Feo
Bonito
Difícil
Fácil
Atractivo
No atractivo
Agradable
Desagradable
Poco importante
Importante
Inútil
Útil
Interesante
Aburrido
RLyLA Vol. 09 (2014), 23-32 | 32
Vol. 9 año 2014, 33-44
EISSN 1886-6298
Approaching lexical loss in Canarian Spanish undergraduates:
A preliminary assessment
Universidad de Las Palmas de Gran Canaria
Abstract: Canarian Spanish is one of the most widely studied varieties of Spanish. Apart from significant distinguishing phonetic
and grammatical features, this variety is notable for a number of characteristics at the lexical level. In their 2009 two-volume
Diccionario ejemplificado de canarismos (henceforth DEC), Corrales and Corbella collected a total of 19,000 entries and 25,000
meanings and expressions, with examples taken from about 1,300 written records. All these Canarianisms represent the history,
natural resources and popular knowledge in the Canarian dialect, thus showing the richness and variety of its vocabulary. Our
aim here is to check whether part of this lexical repertoire is being lost, as previously warned of by Medina (1997). We provide
data obtained in a survey completed by 100 young university students of Canarian origin at the University of Las Palmas de Gran
Canaria on their knowledge and use of a small corpus of Canarianisms taken from the DEC.
Keywords: Sociolinguistics, lexical loss, active and passive vocabulary, Canarian, Spanish.
Language death is a terrible loss, to all who come into contact with it: Facing the loss of
language or culture involves the same stages of grief that one experiences in the process
of death and dying. (Crystal, 2000: 163)
1. introduction
The study of words is one of the most fascinating areas of linguistics. The vocabulary of a language can be
analysed from many and varied perspectives to find out about its origin and meaning, how it changes or how words
relate to one another, but also to get insights on how we use them to view the world and, most importantly, as
markers of regional identity. Many sociolinguists have proved that lexical choices can establish in-group solidarity
and that regional identity can be indexed by lexical items that effectively convey in-group solidarity. According to
Childs and Mallinson (2006: 3), “lexical items may take on marked significance as symbolic vehicles through which
speakers assert and negotiate their ethnic identity”. The reason is that language is recognized as “the primary
index, or symbol, or register of identity” (Crystal, 2000: 40), this concept of identity being defined as “what makes
the members of a community recognizably the same. It is a summation of the characteristics which make it what
it is and not something else – of ‘us’ vs. ‘them’” (Crystal, 2000: 39).
Despite this significant role as identity marker, it is a fact that one of the aspects of a language that may
show rapid change is that of the lexicon. This is hardly surprising since it is the linguistic level which most closely
connects with reality, clearly reflecting all the concepts, entities, objects and changes that take place in the physical
world. Of course, it is nouns and some verbs that change, but not prepositions or pronouns.
In this process of constant change, some words are felt to be old, while others are perceived as new. These
changes in vocabulary do not hamper communication since usually traditional, old words live together with the
new ones for some time. Besides, some new words may not last long; just like some fashions, they may be
very ephemeral. Many new words are introduced into the language with the influence of other languages, new
technologies, novelties in clothing or changes in customs. They may become quite fashionable, and speakers
like to use those words which are modern and prestigious. At the same time, some things may die and bring into
disuse the words required to make reference to them. Similarly, some words are so closely intertwined with cultural
elements that, when the latter evolve and disappear, so do the former. This has already happened to many words
referring to country tools, domestic appliances, old customs or celebrations.
| 33
Approaching lexical loss in canarian spanish undergraduates: a preliminary assessment
Nowadays, a considerable number of words that belong to the realm of our cultural heritage are no longer
used in our urban speech, let alone in the media. They are felt to be old, rustic or provincial, so speakers regard
them with suspicion until they become moribund words (García-Mouton and Grijalbo 2011: 16-17). As Crystal
(2000: 22) remarks, “Knowledge of vocabulary declines, with younger people familiar with only a proportion of
the traditional vocabulary known by older people, and older people being unfamiliar with or antipathetic to the
borrowed vocabulary that is replacing it”.
Within the context of Canarian Spanish, the publication of Corrales and Corbella’s 2009 two-volume dictionary
of Canarianisms placed the enormous richness and variety of Canarian Spanish vocabulary on record. This work
covers a total of 19,000 entries and 25,000 meanings and expressions which are typical of the Canarian lexicon,
with examples taken from about 1,300 written records. All the words collected represent the history, natural
resources and popular knowledge, as well as the past and present of the Canarian dialect. It includes terms used
in popular and educated speech, together with the old and the new voices employed in both rural and urban
areas, words belonging to traditional spheres (such as local sports, agriculture or the maritime context) and also
the new creations promoted by the media. Although other dictionaries have been published before and after the
one in point here, by these and other authors, the Diccionario ejemplificado de canarismos (henceforth DEC) is
considered to be the most comprehensive.
This paper aims at checking whether a small sample of that vocabulary is being lost, as previously warned of
by Medina (1997), by offering a preliminary assessment of the extent to which Canarian university students know
and use words which are typical of their variety. Our article is organised as follows: First, we comment briefly on
the origin and status of Canarian Spanish, devoting a subsection to the role of the pre-hispanic language in the
formation of its lexical repertoire. After referring briefly to some previous empirical studies on the knowledge and
usage of Canarianisms, we describe our initial hypotheses and research procedure. Then, we provide and discuss
the data obtained in a survey completed by 100 young undergraduates of Canarian origin at the Universidad de
Las Palmas de Gran Canaria (henceforth, ULPGC). The paper finishes with some concluding remarks and a few
suggestions for further research.
2. CANARIAN SPANISH: ORIGIN AND STATUS
Canarian Spanish, one of the most widely studied varieties of Spanish (Medina, 1996: 10; Álvarez, 1996: 67;
Corrales, Álvarez and Corbella, 2007 [1988]), is classified among the Atlantic varieties of Spanish, together with
(Western) Andalusian Spanish and the Spanish of the Americas, due to some similarities at the phonetic and
grammatical levels. Despite the many distinguishing features of this variety (Alvar, 1975; Almeida and Díaz-Alayón,
1988; Morera, 1990), here we will be concerned only with its vocabulary, which is characterised by:
•
A number of archaisms from the Castilian Spanish that was used at the time of the conquest.
•
Considerably higher number of loanwords from Portuguese or Galician-Portuguese, Latin American
Spanish, and English than in Castilian Spanish.
•
Some words remaining from the Guanche language spoken by the Guanche aborigines.
The interest of this last feature is undeniable as it is not shared with any other variety of Spanish. Although
most scholars admit the limited influence of the Guanche vocabulary on the formation of the lexical repertoire of
Canarian Spanish, we believe it well deserves the few lines we provide in the subsection below.
Trujillo (1981: 19), Déniz (1996: 330-1) and Morgenthaler (2008: 217) have underlined the fact that there is not a
local prestigious standard variety of Canarian Spanish which could be institutionalised and used as the norm, given
the internal variation that exists in this dialect. However, some studies (Morín and Castellano, 1990; González-Cruz,
2006; Morgenthaler, 2008) suggest that as a non-standard variety, Canarian Spanish seems to have shifted, or be
in the process of doing so, from low to high status. In fact, the study of Canarianisms has attracted growing interest
among researchers in recent years, with the appearance of several compilations of different lengths, besides the
DEC. Among others the following stand out: Lorenzo, Morera and Ortega’s Diccionario de canarismos (1995);
Corrales, Corbella and Álvarez’s Diccionario diferencial del español de Canarias (1996); Corrales and Corbella’s
Diccionario histórico del español de Canarias (2001); Morera’s Diccionario histórico-etimológico del habla canaria
(2006) and Diccionario básico de canarismos (2010) published by the Academia Canaria de la Lengua.
Such academic interest in Canarian Spanish has gone hand in hand with wider social and educational moves to
promote various aspects of Canarian culture (González-Cruz and Vera-Cazorla, 2011: 18-19). Thus, its speakers –
who historically tended to feel their dialectal forms (mainly their lexis and pronunciation) were inferior or inadequate
for usage in formal contexts or in the presence of Castilian Spanish speakers (Ortega, 1981) – seem nowadays to
RLyLA Vol. 09 (2014), 33-44 | 34
be showing some pride in Canarian Spanish and to understand that it deserves respect (Morín and Castellano,
1990; Morín, 1993; Almeida, 1994; Hernández, 2003; Morera, 2002), as long as it constitutes an important part of
their intangible cultural heritage. However, as a dynamic, living dialect it still suffers the threat of various types of
linguistic generalizations, that is, of processes of convergence towards the standard forms, possibly through the
pressures of the Media. In line with Dalby’s (2002: 256) ideas, we believe that not only every language, but also
every dialect “offers its own classification of living things, natural phenomena and cultural concepts”; therefore,
Canarian words represent “a different way of looking at, mapping and classifying the world” (272), hence, it is
important to keep them alive.
The Guanche element
Many mysteries and uncertainties surround the history and origin of the Guanches, the primitive people living
in the Canaries before the European conquest. Although there are several theories, most scholars seem to agree
that they were closely related to the North African Berbers. After their incorporation into the Crown of Castile in
1496, the islands were settled by Spanish-speaking colonists from the continent, who imposed their language and
culture on the native population. Although the sources indicate that they survived the conquest in considerable
numbers, after a brief period of bilingualism the natives adopted the Castilian language and abandoned their own.
Regarding the pre-hispanic influence on Canarian Spanish, many authors, local and foreign, have contributed
to their study (Glas, 1764; Bute, 1891; Abercromby, 1917; Álvarez-Rixo, 1991; Wölfel, 1996/1965; Trapero and
Llamas, 1998; Trapero, 1999, 2007, among others). There are some references to the Guanche language in the
historic and literary primary sources written soon after the Castilian conquest, but it is not until the second half
of the XIXth century that proper linguistic studies begin to try to identify and characterise this dead language.
Despite the many pages published on the topic, the study of the Guanche language is still to be completed,
although a global description is obviously an impossible task, as Trapero (2007: 119) explains, since no phonetic
or grammatical evidence remains, the only approachable domain being that of the lexicon.
Research on the Guanche terms is certainly interesting since it is what makes Canarian Spanish lexicon
different from the vocabulary of general Spanish. Díaz-Alayon (1991: 54) distinguishes the following three channels
through which the pre-hispanic lexical component has remained and been transmitted up to the present. Firstly,
the historical and administrative sources have to be considered. Obviously, in their accounts each scribe followed
his own criteria and tended to collect a different number of Guanche terms, which were probably more or less
Spanishized in their transcription. Secondly, some Guanche words have survived to a greater or lesser extent in
the language substratum. Thirdly, we must mention the large group of toponyms that remain all over the islands.
Samper (2008: 164) believes that the Guanches’ influence is limited to “a short list of words (approximately
120 today, according to Corbella, 1996) that, aside from those that survive in toponymy and anthroponymy, refer
principally to flora, the terrain, fauna, and the worlds of agriculture and animal husbandry”. More accurately, when
comparing the incorporation of indigenous words into the Spanish used in the contexts of the American New
World and the Canaries, Bravo-García and Cáceres-Lorenzo (2011: 105-7) underline the lack of prestige of the
Guanche language, which might explain why only 25 Guanche terms were used in the official documents written
between the times of the Conquest and 1550. They constitute a mere 17.69% of the total number of words of
Guanche origin which are known nowadays, amounting to a total of 147, according to the records in the historical
dictionaries. This contrasts with the higher number of words of Portuguese origin (128 up to 1550, and 1200 up
to the present), which indicates the competition between the two languages in the formation of early Canarian
Spanish lexical repertoire. This little information on the Guanche language and culture seems to suggest that the
natives were treated as outcasts and that they experienced acculturation to a considerable degree (CáceresLorenzo, 2007: 50).
Trapero (1999; 2007), who has studied the issue thoroughly, is conscious of the complex problems and
drawbacks that surround the description of the Guanche language and explains them neatly (1999: 20-29).
He complains that very few authors have based their research on the oral tradition, which is – he argues – the
only primary source for linguistic issues and the only true one for living toponymy (2007: 52). In fact, he states
that within the Guanche lexical elements, toponyms constitute 90% (Trapero, 2007: 111). Thus, although the
Guanche language was lost very soon (two or three, at the most, generations after the Conquest), the Canaries
have maintained a considerably wide and representative aboriginal toponymic corpus (Trapero, 2007: 20). This
author is currently preparing a dictionary of Guanche toponymy (Diccionario de toponimia guanche). His 2007
work includes a provisional catalogue of Guanche terms currently present in the islands’ toponymy, a collection of
1,803 lexical units, some of them with variants. In his opinion, any future study on the Guanche language must be
complemented with the data obtained by scholars working in other disciplines, such as Prehistory, Archaeology or
Anthropology. This will be helpful to shed more light on and determine the origin of the Guanches, the chronology
of their arrival and the differences in their culture and identity.
RLyLA Vol. 09 (2014), 33-44 | 35
Before describing our own research, we will refer briefly below to the valuable contributions of other authors
who have investigated, more or less extensively, the vitality of certain Canarianisms in Canarian people’s presentday speech.
3. LITERATURE REVIEW
Several specialists have tested empirically the knowledge and usage of Canarianisms by today’s Canarians.
Medina (1997) surveyed a group of 30 university students of Spanish Philology at the University of La Laguna
(Tenerife) to find out whether they knew a list of 60 Guanche terms, as well as their opinions about this type of
word. He concludes that a considerable number of the words listed were unknown and poorly defined. Only two of
the pre-Hispanic terms, baifo (‘kid’) and gofio (‘a typical flour made of toasted corn and/wheat’), were recognised
by all the informants, followed by tabaiba (‘a bush’) and perenquén (‘a wall gecko’), which were known by 96% of
the students, and mago (‘countryman’) by 93.3%. Most students confirmed their having heard some time many of
the terms they were unable to define, especially from old country people. They were well aware of the fact that all
this vocabulary was part of the rich Canarian ethnographic and linguistic legacy, but they considered it to be no
longer in use or very restricted (Medina, 1997: 797).
When studying the Canarianisms collected by the famous Canarian writer Pérez-Galdós, Hernández and
Samper (2003) included the results of their own research on the vitality of 174 of those words. They surveyed 40
informants from Las Palmas de Gran Canaria, belonging to two different generational groups, with the following
distribution:
•
10 men and 10 women with ages ranging between 25 and 35 (1st generation).
•
10 men and 10 women who were older than 55 (3rd generation).
Since previous studies have proved that informants with a low socioeconomic level tend to keep in use many
words which are considered to be archaic, Hernández and Samper (2003) selected informants with a low profile in
education (only primary studies) and in their professions (blue-collar jobs with a limited income). They differentiate
between the active and the passive vocabulary in the informants’ lexical repertoire and provide an index of lexical
loss for each of the lexical units in their survey corpus: 39.6% (i.e., 69 out of the total of 174 lexical units) of the
Canarianisms in their corpus remain active or living in the informants’ speech, with 7 terms being known and used
by all of them. It is interesting to note –as these authors do– that 4 of the 69 words in this group of high vitality are
Guanche terms, namely perinquén, baifo, beletén and tabaiba.
The other side of the coin is represented by the 105 Canarianisms which are known or used by less than 50%
of the informants. Specifically, there are 6 words which nobody knows or uses, so they have already disappeared
from the informants’ lexical repertoire, while 10 are still known but never or only very rarely used. Factors like
age and gender are analysed, the main conclusions being that the former is especially relevant to explain the
differences in knowledge and usage among the two generational groups of informants, while the latter is not.
Interestingly, men tend to know more terms related to agriculture and the sea than women, which can easily be
explained by the different social roles traditionally played by each gender.
Although Canarianisms is not their focus, other works worth-mentioning here are those published by the
Canarian Government (1986), Samper (1998) and Samper-Hernández (2009). In 1986 the Department of Education
of the Canarian Government edited their Léxico del español usual en Canarias, an inventory of about 2000 words
of frequent usage in the islands. This lexical repertoire included some archaisms, a few pre-hispanic terms, some
Americanisms and Portuguesisms and other words of various origins, all of them considered to be of special use
in the Canaries. Among the Canarianisms registered we found three that were used in our questionnaire: frangollo,
gánigo, and desmayarse, the latter with its special meaning ‘to yawn’.
Samper (1998) participated in an international Project, coordinated by Lope-Blanch, to study the rules for
educated speech in the main cities of the Spanish-speaking world. Interestingly, out of the 15,506 different
responses obtained with a questionnaire based on 21 semantic fields, only one Guanchism was mentioned,
namely, perenquén (precisely the Canarianism most widely known and frequently used by our informants, as we
will see below). What they find remarkable about this word is the number of variants it has (peninqué, perinquén,
perenqué, peninquén, perinqué).
More recently, Samper-Hernández (2009) studied the evolution of the available lexicon in primary and
secondary students from Gran Canaria. She found that 5% of the total number of terms given by her informants
could be classified as Canarianisms, their semantic fields being, in decreasing order, those related to ‘food and
drinks’, ‘the country’, ‘animals’ and ‘the kitchen and cooking utensils’. These domains seem to be closely linked
RLyLA Vol. 09 (2014), 33-44 | 36
to the environment and local traditions, and therefore tend to favour the presence of dialectal words. Interestingly,
80 Canarianisms appeared in the first 100 positions, the conclusion being that they are highly available. On the
other hand, the correlation between the informants’ sociocultural level and their usage of Canarianisms was also
confirmed in this research: speakers with the lowest sociolect tend to use a higher number of Canarianisms,
while those with the highest sociolect showed remarkable, almost encyclopedic, knowledge of words referring to
the natural environment, by giving a higher number of Canarianisms related to the domains of ‘the country’ and
‘animals’. Therefore, Samper-Hernández (2009:402) concludes that speakers of the lowest social level seem to be
more familiar with this type of word than those of a higher status.
4. RESEARCH PROCEDURE AND MATERIALS
Conceived as a cross-sectional and therefore descriptive study, our research began with the following
hypotheses:
1. We believed that a considerable number of the Canarianisms registered in the DEC would be unknown
and rarely used by our informants (a particular section of the Canarian population, that of university
youngsters) not only because many of them have to do with items of the Canarian natural environment
(plants and animals) but because a large majority of the lexical units belong to the popular language used
in the rural world, refer to country traditions or constitute a very informal and popular vocabulary which is
associated with the elderly and is not used by the more modern, younger and more globalised sectors of
the population.
2. We also assumed that since women have a reputation of being more conservative (or less innovative than
men) regarding language usage (Trudgill, 1985; López-Morales, 1989; Holmes, 1992; Romaine, 1994;
Almeida, 1999; Wardhaugh, 2002), perhaps our female informants would know and use a higher number
of Canarianisms. Besides, women are said to have a wider lexical repertoire than men (Morín, 1993: 93).
Firstly, we decided to randomly select a sample of the lexical items from the wide dictionary corpus. Five
ULPGC teachers of Canarian origin and ages ranging between 45 and 62 helped us in the selection process. They
opened each of the two dictionary volumes four times at different points, to choose one item on a random basis,
but trying to favour those words they themselves knew. Thus we obtained a first round of 40 words. Since we did
not want our survey to be too long and tiring for the informants, we carried out a further selection to reduce the
total number of words to 201. For this we just wrote each of the initial 40 words in a strip of paper, which we folded
and put in a bag. Finally, 20 pieces of paper were blindly taken out of the bag by the researcher, thus obtaining our
sample of 20 Canarianisms (see Table 1 below). Interestingly, eight of these terms are of Guanche origin, namely,
beletén, gánigo, guanil, jairo, mocán, perenquén, tabona and tajaraste. The point is that though comparatively
scarce, these Guanche terms are relatively popular.
Although some of the Canarianisms in our survey corpus can have several meanings, in general the words
selected can be broadly classified as belonging to the following semantic areas:
* Animals: bocinegro, capirote, jairo, perenquén
* Plants and trees: mocán, oreja de burro, pico de paloma, viñátigo, yerba de vidrio
* Natural world: piedra viva, tabona
* Words related to
cultural traditions: beletén, botana, gánigo, guanil, frangollo, tajaraste
* Popular verbs and
adjectives: 1
chirgarse, esmayarse, desmandarriado
Admittedly, our sample of 20 randomly-selected words can hardly be considered as representative of a corpus of 19,000 lexical units, as they just mean a mere 0.10%
of the total. However, our study is a preliminary one, not intended to be exhaustive. Similarly, we justify the number of informants by resorting to Labov’s (1966: 170-71)
claim that when dealing with linguistic behavior it is only necessary to study a 0.025% of the whole population. Our sample of 100 informants constitutes 0.5% of the
total number of students at ULPGC (n= 20,000), therefore it is representative.
RLyLA Vol. 09 (2014), 33-44 | 37
Table 1 below provides a description of the meanings of our sample of Canarianisms2, as well as possible
equivalents (when there is one) in general Spanish.
Table 1. Meanings and equivalents in general Spanish for the 20 Canarianisms in the survey.
CANARIANISMS
1. BELETÉN
2. BOCINEGRO
3. BOTANA
4. CAPIROTE
5. CHIRGARSE
6. DESMANDARRIADO
7. ESMAYARSE
8. FRANGOLLO
9. GÁNIGO
10. GUANIL
11. JAIRO
12. MOCÁN
13. OREJA DE BURRO
14. PERENQUÉN
15. PICO DE PALOMA
16. PIEDRA VIVA
17. TABONA
18. TAJARASTE
19. VIÑÁTIGO
20. YERBA DE VIDRIO
MEANINGS
‘Colostrum, the first milk produced by goats, sheep or
cows after giving birth’
‘A fish’
‘A leather case to cover a fighting cock’s spur’
‘A bird’
‘To get frightened’
‘Dishevelled, dressed in a slovenly way’
‘To be starving / to yawn’
‘A typical dessert made of corn’
‘A clay pot’
‘An unmarked animal’
‘ A billy-goat’
‘A tree’
‘A garden plant’
‘A wall gecko’
‘A plant’
‘A very hard basalt stone’
‘A Guanche knife / an obsidian stone’
‘A typical dance / a tambourine’
‘A tree and/ or its wood’
‘A plant’
General Spanish
Calostro
Asustarse
Desaliñado
Tener mucha hambre / bostezar
Vasija de barro
Macho cabrío
Salamandra, lagartija
With this material we designed a very concise chart so that our informants could supply in it information about
their knowledge and use of each of these terms. What we wanted to know was simply
a. whether they knew the meaning or at least a general referent for each Canarianism,
b. whether they ever used them and with which frequency (never, sometimes or often),
c. what type of speakers they thought would probably use those words,
d. how they felt about their (not) knowing those words.
This short questionnaire was distributed in two of the ULPGC campuses (Tafira and Obelisco) among 100
students of various specialities in Sciences and Humanities, respectively, between September and October
2012. Our informants were 50 female and 50 male students with ages ranging between 19 and 24. Most of the
female participants (43) were students at the faculty of Philology, while only 11 male informants studied there. The
rest of the participants studied at the faculty of Economics (7 female and 18 male). All these students belonged
to intact groups whose teachers kindly gave us permission to survey. Informants participated voluntarily, the
only requirement for them being to have Canarian origin and upbringing. The rest of the male participants were
contacted individually or in small groups in the Tafira campus and were students at the faculties of Engineering
(12) and Computer Science (9).
5. RESULTS AND DISCUSSION
The tables below summarise the main results obtained in our survey. Table 2 shows global results regarding our
informants’ correct or relatively correct knowledge of the meaning of the proposed Canarianisms and the usage
they make of them. Table 3 provides more specific data, considering our informants’ responses by gender, as
2
It is important to give here the definition of the concept of Canarianism used by the DEC’s authors, who refer to any “word, phrase or way of speaking peculiar to the
inhabitants of the Canaries, whose use has settled into both the spoken and written language.” What Corrales and Corbella (2009: xvi) mean by peculiar is: “not only
words exclusive to the Canaries but also those used in Spanish-speaking areas –paying attention to the history and geographical position of the archipelago, meeting
point for the words that circulated around the Atlantic– the use or different labelling of the words (the more frequent use of hereditary terms or of certain variants, the
continued use of words considered archaic in standard Spanish, variation or specialisation of a meaning, the adaptation of nautical or agricultural terms to the everyday
language, the different diastratic or diaphasic updating for some words the different syntactic combination, the abundant lexicalisation of diminutives, changes in grammatical categories, etc.)” (Our translation).
RLyLA Vol. 09 (2014), 33-44 | 38
well as by degree of knowledge. We specify if they are able to provide a more or less complete description of the
meaning of the word or if they simply give a very general referent; likewise, we make a distinction in their frequency
of usage, that is, whether they use the word very often or just sometimes.
The most interesting results are those in Table 4, which contrasts the terms known and used, forming the
active vocabulary, and those that are simply known but never used, thus forming the passive vocabulary. The latter
represent the initial step which may lead a word towards disappearance or loss (Hernández and Samper, 2003: 42).
These data allowed us to establish the percentage or degree of lexical loss for each word.
Table 2. Global results regarding general knowledge of meaning and usage.
CANARIANISMS
1. BELETÉN
2. BOCINEGRO
3. BOTANA
4. CAPIROTE
5. CHIRGARSE
6. DESMANDARRIADO
7. ESMAYARSE
8. FRANGOLLO
9. GÁNIGO
10. GUANIL
11. JAIRO
12. MOCÁN
13. OREJA DE BURRO
14. PERENQUÉN
15. PICO DE PALOMA
16. PIEDRA VIVA
17. TABONA
18. TAJARASTE
19. VIÑÁTIGO
20. YERBA DE VIDRIO
Number of
informants who
give a relatively
correct meaning
28
24
0
19
2
33
27
52
1
2
10
12
1
91
3
4
1
11
14
1
Number of
informants who
give a wrong
meaning
1
1
1
22
10
12
16
3
2
2
13
1
6
4
4
8
2
20
5
0
Percentage of
informants who
do not know
the meaning
72%
75%
100%
81%
98%
67%
73%
48%
99%
98%
90%
88%
99%
9%
97%
96%
99%
89%
86%
99%
Number of
informants
who ever use
the word
15
10
0
8
1
23
17
17
1
1
8
4
0
76
1
1
0
7
7
0
Percentage of
informants who
NEVER use the
word
85%
90%
100%
92%
99%
77%
83%
83%
99%
99%
92%
96%
100%
24%
99%
99%
100%
93%
93%
100%
Table 2 above shows that only two Canarianisms stand out as being really well known by more than 50% of the
informants; they are perenquén (91) and frangollo (52), although the former is much more often used (76%) than
the latter (17%). They are followed, in decreasing order of degree of knowledge, by the following: desmandarriado
(33), beletén (28) and esmayarse (27).
Interestingly, a considerable number of informants give a wrong meaning for the following Canarianisms:
capirote (22), tajaraste (20), esmayarse (16), jairo (13) and desmandarriado (12). It is also noticeable that only seven
of the Canarianisms in our survey corpus have an equivalent in general Spanish. In this group, only perenquén
seems to be favoured versus the more general terms salamandra/ lagartija.
Equally outstanding is the case of those terms which none (100%) or almost none (99%) of our informants
use, such as botana, tabona and yerba de vidrio (100%), and chirgarse, guanil, pico de paloma and piedra viva,
which just have one user (99%). Another remarkable observation that can be made has to do with the number
of informants who give a wrong meaning for some of the terms. This is the case of capirote (22), tajaraste (20),
esmayarse (16), jairo (13), desmandarriado (12) and chirgarse (10). In most of them, the reasons for the errors are
relatively easy to understand. Thus, capirote is used in general Spanish to refer to a very sharp-pointed hat that
is typically worn by the participants in religious processions at Easter. The similarity in form between the two
Guanche words tajaraste and tajinaste, explains why many informants erroneously gave the latter’s meaning (‘a
plant’) for the former, which refers to ‘a dance’. Likewise, esmayarse (‘to be hungry/to yawn’) was confused with
desmayarse (‘to faint’) and chirgarse (‘to get frightened’) with chingarse (‘to splash yourself’). Jairo (‘a billy-goat’)
was mistaken as a proper name since in fact there is a Latin-American singer called Jairo, who used to be quite
popular some years ago. Finally, Table 3 below shows more specific data regarding degree of knowledge and
frequency of usage, both classified by gender.
RLyLA Vol. 09 (2014), 33-44 | 39
Table 3. Results by gender and degree of knowledge and usage.
CANARIANISMS
1. BELETÉN
2. BOCINEGRO
3. BOTANA
4. CAPIROTE
5. CHIRGARSE
6. DESMANDARRIADO
7. ESMAYARSE
8. FRANGOLLO
9. GÁNIGO
10. GUANIL
11. JAIRO
12. MOCÁN
13. OREJA DE BURRO
14. PERENQUÉN
15. PICO DE PALOMA
16. PIEDRA VIVA
17. TABONA
18. TAJARASTE
19. VIÑÁTIGO
20. YERBA DE VIDRIO
Number of men and
women who give
a complete and
correct meaning
M
F
10
5
19
5
0
0
13
4
1
1
13
15
12
15
25
25
1
0
2
0
4
5
8
1
1
0
45
46
3
0
2
0
1
0
5
5
3
2
1
0
Number of men and
women who give
a partially correct
meaning
M
F
6
7
0
0
0
0
2
0
0
0
4
1
0
0
1
1
0
0
0
0
0
1
1
2
0
0
0
0
0
0
2
0
0
0
0
1
5
4
0
0
Number of men
and women who
use the word
frequently
M
F
0
1
2
0
0
0
1
1
0
0
3
6
2
5
1
1
0
0
0
0
2
3
1
0
0
0
11
17
1
0
1
0
0
0
0
3
1
0
0
0
Number of men
and women who
use the word
sometimes
M
F
9
5
4
4
0
0
5
1
0
1
7
7
4
6
8
7
1
0
1
0
1
2
1
2
0
0
22
26
0
0
0
0
0
0
2
2
5
1
0
0
Table 3 above shows more men than women give complete meanings for 12 words, their scores being remarkably
different in four of them, namely beletén, bocinegro, capirote and mocán. Women only slightly surpass men in their
knowledge of four words: esmayarse, jairo, perenquén and tajaraste (if we sum the number of complete and partially
correct answers). In turn, Table 4 below shows the extent to which each of the Canarianisms studied belongs to our
informants’ active or passive vocabulary, as well as the corresponding index of lexical loss for each word.
Table 4. Percentages of active vocabulary, passive vocabulary and lexical loss.
CANARIANISMS
1. PERENQUÉN
2. DESMANDARRIADO
3. FRANGOLLO
4. ESMAYARSE
5. BELETÉN
6. BOCINEGRO
7. CAPIROTE
8. VIÑÁTIGO
9. TAJARASTE
10. JAIRO
11. MOCÁN
12. PIEDRA VIVA
13. PICO DE PALOMA
14. CHIRGARSE
15. GUANIL
16. GÁNIGO
17. OREJA DE BURRO
18. TABONA
19. YERBA DE VIDRIO
20. BOTANA
ACTIVE VOCABULARY
(words known and used)
(M+F)
Total
(33 + 43) 76
(10 + 13) 23
(9 + 8)
17
(6 + 10)
16
(9 + 6)
15
(6 + 4)
10
(6 + 2)
8
(6 + 1)
7
(2 + 5)
7
(3 + 5)
8
(2 + 2)
4
(1 + 0)
1
(1 + 0)
1
(0 + 1)
1
(1 + 0)
1
(1 + 0)
1
(0 + 0)
0
(0 + 0)
0
(0 + 0)
0
(0 + 0)
0
PASSIVE VOCABULARY
(words known but not used)
(M+F)
Total
(12 + 3 )
15
(7 + 3)
10
(16 + 19)
35
(6 + 6)
12
(7 + 7)
14
(14 + 0)
14
(7 + 2)
9
(2 + 5)
7
(3 + 1)
4
(1 + 1)
2
(7 + 1)
8
(3 + 0)
3
(2 + 0)
2
(1 + 0)
1
(1 + 0)
1
(0 + 0)
0
(1 + 0)
1
(1 + 0)
1
(1 + 0)
1
(0 + 0)
0
PERCENTAGE OF
LEXICAL LOSS
Total
9%
67%
48%
72%
71%
76%
83%
86%
89%
90%
88%
96%
97%
98%
98%
99%
99%
99%
99%
100%
The figures above indicate that only one word, perenquén, is safely registered in our informants’ lexical repertoire,
reaching the highest percentage (76%) within the active vocabulary and the lowest index of lexical loss (9%). It is
followed at great distance by desmandarriado and frangollo, which seem to compete for second position in our
ranking since, although the former is more actively known and used (23%), the latter scores higher (35%) in the
passive vocabulary. This means that frangollo obtains a lower index of lexical loss (48%) than desmandarriado (67%).
RLyLA Vol. 09 (2014), 33-44 | 40
The next three words in the list (esmayarse, beletén and bocinegro) are still slightly known and used by our
informants, with their lexical-loss index in the 70s. It is obvious that the next five terms (capirote, viñátigo, tajaraste,
jairo and mocán) are in a very dangerous situation, with only about ten of our informants knowing and using them.
The case of the following words is also clear: except for piedra viva and pico de paloma, recognized respectively
by merely 4 and 3 informants, the rest of the Canarianisms in the list are just known by one informant, botana being
utterly unkown.
Regarding the third question in our survey (what type of speakers would probably use those words according
to our informants) we must say that not all students gave an answer, neither referred to all the words. The reason
is possibly that they were not familiar with those terms, so they did not know who would use them. Table 5 below
summarises our informants’ responses to this question (some answers included more than one option).
Table 5. Informants’ opinions about who would use each of the Canarianisms.
Who uses it? 
CANARIANISMS
1. BELETÉN
2. BOCINEGRO
3. BOTANA
4. CAPIROTE
5. CHIRGARSE
6. DESMANDARRIADO
7. ESMAYARSE
8. FRANGOLLO
9. GÁNIGO
10. GUANIL
11. JAIRO
12. MOCÁN
13. OREJA DE BURRO
14. PERENQUÉN
15. PICO DE PALOMA
16. PIEDRA VIVA
17. TABONA
18. TAJARASTE
19. VIÑÁTIGO
20. YERBA DE VIDRIO
Do not know
72
92
92
69
89
65
65
62
91
87
84
85
87
29
84
85
89
77
82
92
Specialists
in the field /
Anyone/all Canarians Country people Older people professionals
4
13
11
0
5
5
5
13
0
3
5
0
7
9
15
0
3
3
5
0
18
5
12
0
10
7
18
0
10
8
13
7
1
3
5
0
0
4
9
0
2
5
8
1
1
5
6
3
0
3
8
2
37
17
17
0
0
5
5
6
3
4
7
1
2
3
5
1
7
3
6
7
4
7
5
2
0
3
5
0
As can be observed, a large majority of our informants seem to have very little or no idea about who would use
the Canarianisms in the survey. In a few cases, it is only a small percentage (always less than 20%) of informants
that mention older people (e.g., esmayarse, 18%), or country people (perenquén, 17%) as possible users. Once
again, perenquén is the Canarianism that obtains the highest percentage (37%) of informants who believe it can be
used by anyone, i.e., by all Canarians. Likewise, the word that obtains the highest percentage of informants who
believe it can be used by specialists is bocinegro (with just a 13%).
Last, but not least, the fourth question brings to the fore the issue of the cultural and social significance of
lexical choices, and tries to determine the extent to which our informants feel these and other Canarianisms play
a role in the construction of regional identity and in-group solidarity. In short, what we wanted to find out was
how our informants feel after checking that they (do not) know (many of) these Canarianism, and whether these
undergraduates actually felt the emotions of loss mention by Crystal in the epigraph quoted above. Table 6 below
summarises their answers:
Table 6. Our informants’ feelings about their (not) knowing the selected Canarianisms.
FEELINGS
It’s a real pity/shame that they do not know the words and they may be lost
Relatively happy/proud about knowing at least some
Curious about the meanings of the words they do not know
Not worried at all for not knowing (some of) the words
Surprised at not having heard/seen some/many of the words
Losing the words of another generation is normal
Nothing, no feeling at all
No answer
N/%
58
3
6
3
9
6
12
3
RLyLA Vol. 09 (2014), 33-44 | 41
As can be observed, 58% of the students –a relatively large though not overwhelming majority– feel ashamed
and concerned that they do not know or use these words and that they are disappearing, simply because they
believe they constitute an important part of their Canarian identity, which is something “they feel proud of”. Some
complained that in their vocabulary many words were being replaced by more modern ones or even foreign words.
This seems to reveal that they are aware of the significance of preserving and not losing this lexical legacy, which
confirms to a considerable extent the reality of their “emotions of loss” as referred to by Crystal.
6. CONCLUDING REMARKS
The data obtained in this survey seem to indicate that our first hypothesis was correct: a majority of the
Canarianisms in our small survey corpus is rather unknown and rarely or never used by our informants, with
just a few exceptions. Only one word, perenquén, seems to be firmly established in the list of active vocabulary
(being known and used by 76% of our informants), and it is followed at great distance by desmandarriado (23),
frangollo (17), esmayarse (16), beletén (15) and bocinegro (10). The remaining terms are only known and used by
less than ten of the informants. One word, botana, has already disappeared from our informants’ lexical repertoire
since none of them know it or use it, while other seven words stand in the lowest levels of the ranking with merely
two or fewer informants who know them and occasionally use them, or simply have them in their passive lexical
repertoire.
The findings also show that our second hypothesis cannot be validated since women only give more completely
or partially correct meanings than men for just four words: esmayarse, jairo, perenquén and tajaraste (see Table 3).
In the light of these data, it seems obvious that a majority of the words in this particular corpus of Canarianisms
constitute lexical material that tends to disappear (Medina, 1997: 796). In the long run, most of it will probably end
up being merely passive vocabulary that only a minority of Canarian Spanish speakers will be able to recognize.
This preliminary assessment of our undergraduates’ knowledge and usage of Canarianisms is therefore negative:
they seem to be currently suffering considerable lexical loss. When they realize this is the case, most of them feel
sorry and concerned about it and say it is important to maintain this vocabulary which provides them and their
region with their own cultural identity.
Notwithstanding, further research with larger samples of both Canarianisms and informants is probably needed
to confirm these preliminary results. Following Morgenthaler’s (2002) line of thought, it will also be of interest to
investigate the extent to which the current process of globalization is affecting the cultural and linguistic identity of
our Canarian youngsters. They are the new generation of speakers of Canarian Spanish and their role in maintaining
or changing the future of this variety is crucial.
A final question to pose at this stage is “what can be done to prevent this and further lexical loss?” Unfortunately
we are not in a position to give a clear answer, since this raises a host of issues which have not been discussed
here. As Crystal (2000: 127-8) admits, when dealing with language death –which we believe can also be applied
to lexical loss– “clusters of factors interact in subtle ways”. Using his words, we cannot but conclude as follows:
There is still so much that we don’t know. What motivates the members of a community to work for their
language? Why do some communities become so involved and others do not? […]. Nor is the range of factors
and how they interact completely understood. We know a great deal about why languages become endangered
and die, and why people shift from one language to another, but we still know very little about why they are
maintained, and why people stay loyal to them.
REFERENCES:
Abercromby, J. (1917). “A study of the ancient speech of the Canary Islands”, Harvard African Studies 1: 95-129.
Academia Canaria de la Lengua. (2010). Diccionario básico de canarismos. Tenerife: Academia Canaria de la
Lengua / Gobierno de Canarias.
Almeida, M. (1994). “Creencias y actitudes lingüísticas en el español canario”, Anuario de Lingüística Hispánica
10: 9-23.
Almeida, M. (1999). Sociolingüística. La Laguna: Servicio de Publicaciones de la Universidad de La Laguna.
Almeida, M. and Díaz-Alayón, C. (1988). El español de Canarias. Santa Cruz de Tenerife: Litografía A. Romero.
Alvar, M. (1975) Atlas lingüístico y etnográfico de las Islas Canarias. Las Palmas de Gran Canaria: Cabildo Insular
de Gran Canaria.
RLyLA Vol. 09 (2014), 33-44 | 42
Álvarez, M.A. (1996). “Sobre algunas construcciones sintácticas en el español de Canarias”, in J. Medina and D.
Corbella (eds.), El español de Canarias hoy: análisis y perspectivas. Frankfurt and Main: Vervuert / Madrid:
Iberoamericana. pp. 67-82
Álvarez-Rixo, J.A. (1991). Lenguaje de los antiguos isleños (Ed. and notes by C. Díaz-Alayón and A. Tejera-Gaspar).
Santa Cruz de Tenerife: Excmo. Ayuntamiento del Puerto de la Cruz/Centro de la Cultura Popular Canaria.
Bravo-García, E. and Cáceres-Lorenzo, M.T. (2011). La incorporación del indigenismo léxico en los contextos
comunicativos canario y americano (1492-1550). Bern: Peter Lang.
Bute, J.P.C. (1891). On the ancient language of the natives of Tenerife. London: J. Masters and Co.
Cáceres-Lorenzo, M.T. (2007). “New contributions to language contact in the Atlantic: the Lascasian view on the
Canaries”, Crítica Hispánica 29/1-2: 47–61.
Childs, B. and Mallinson, C. (2006). “The Significance of Lexical Items in the Construction of Ethnolinguistic
Identity: A Case Study of Adolescent Spoken and Online Language”, American Speech, 81/1: 3-30.
doi:10.1215/00031283-2006-001
Corbella, D. (1996). “Fuentes del vocabulario canario: los préstamos léxicos”, in J. Medina and Corbella, D. (eds.),
El español de Canarias hoy: análisis y perspectivas. Frankfurt and Main: Vervuert / Madrid: Iberoamericana.
pp. 105-141.
Corrales, C, Corbella, D. and Álvarez, M.A. (1996). Diccionario diferencial del español de Canarias. Madrid: Arco
Libros, S. L.
Corrales, C. and Corbella, D. (2001). Diccionario histórico del español de Canarias. La Laguna: Instituto de Estudios
Canarios.
Corrales, C. and Corbella, D. (2009). Diccionario ejemplificado de canarismos. La Laguna: Instituto de Estudios
Canarios, 2 vols.
Corrales, C., Álvarez. M.A. and Corbella, D. (2007 [1988]). El español de Canarias. Guía bibliográfica. Tenerife:
Instituto de Estudios Canarios.
Crystal, D. (2000). Language Death. Cambridge: Cambridge University Press. doi:10.1017/CBO9781139106856
Dalby, A. (2002). Language in Danger. London: Allen Lane.
Déniz, F. (1996). “Las hablas canarias y el poder”. El Museo Canario, 51: 321-336.
Díaz-Alayón, C. (1991). “La contribución léxica de las lenguas prehispánicas al español de Canarias”, Estudios
filológicos 26: 51-69.
García-Mouton, P. and Grijelmo, A. (2011). Palabras moribundas. Madrid: Taurus.
Glas, G. (1764). The history of the discovery and conquest of the Canary Islands, translated from a Spanish
manuscript lately found in the island of Palma. London: R. and J. Dodsley, and T. Durham.
González-Cruz, M.I. (2006). “Subjective Reactions to Two Spanish Accents: A Sociolinguistic Survey of ULPGC
Students”, in M.I. González-Cruz (coord.) Lengua, sociedad y cultura: estudios interdisciplinares. Las
Palmas de Gran Canaria: ULPGC, Servicio de Publicaciones / La Caja de Canarias. Obra Social. pp. 53-77.
González-Cruz, M.I. and Vera-Cazorla, M.J. (2011). “The Construction of Canarian Identity in the First Telenovela
on Canarian Television”, in N. Lorenzo-Dus (ed.) Spanish at Work. Analysing Institutional Discourse across
the Spanish-Speaking World. Basingstoke, England: Palgrave MacMillan. pp. 16-34.
Hernández, H. (2003). “Dialectalismos, medios de comunicación y enseñanza de la lengua”, Estudios sobre el
español de Canarias. Actas del I Congreso Internacional sobre el español de Canarias, vol. 2. Islas Canarias:
Litografía Romero. pp. 895-909.
Hernández, C.E. and Samper, J.A. (2003). Voces canarias recopiladas por Galdós. Las Palmas: Cabildo de Gran
Canaria.
Holmes, J. (1992). An Introduction to Sociolinguistics. New York: Longman.
Labov, W. (1966). The Social Stratification of English in New York City. Washington, D.C.: Center for Applied
Linguistics.
Léxico del español usual en Canarias (1986). Gobierno de Canarias / Consejería de Educación.
López-Morales, H. (1989). Sociolingüística. Madrid: Gredos.
Lorenzo, A., Morera, M. and Ortega, G. (1995). Diccionario de canarismos. La Laguna: Francisco Lemus Editor.
Medina, J. (1996). “La investigación lingüística sobre el español de Canarias”, in J. Medina and D. Corbella (eds.)
El español de Canarias hoy: análisis y perspectivas. Frankfurt and Main: Vervuert / Madrid: Iberoamericana.
pp. 9-48.
Medina, J. (1997). “Los prehispanismos canarios: quinientos años después”, in J.L. Otal, I. Fortanet and V. Codina
(eds.) Estudios de lingüística aplicada. Castelló de la Plana: Universitat Jaume I. pp. 793-799.
Morera, M. (1990). Lengua y colonia en Canarias. La Laguna: Editorial Globo.
Morera, M. (2002). “La identidad lingüística del canario”, Disenso. Revista Canaria de análisis y opinión 32: 10-12.
RLyLA Vol. 09 (2014), 33-44 | 43
Morera, M. (2006). Diccionario histórico-etimológico del habla canaria. Cabildo de Fuerteventura. Servicio de
Publicaciones.
Morgenthaler, L. (2002). “Globalización lingüística y pérdida de identidad”, Disenso. Revista Canaria de análisis y
opinión, 38: 7-9.
Morgenthaler, L. (2008). Identidad y pluricentrismo lingüístico. Hablantes canarios frente a la estandarización.
Madrid / Frankfurt and Main: Vervuert / Iberoamericana.
Morín, A. and Castellano, A. (1990). “Creencias y actitudes del docente canario hacia su modalidad de habla”, El
Guiniguada, 1: 257-261.
Morín, A. (1993). Actitudes sociolingüísticas en el léxico de Vegueta. Servicio de Publicaciones de la Universidad
de Las Palmas de Gran Canaria.
Ortega, G. (1981). “El español hablado en Canarias: visión sociolingüística”, Revista de Filología de la Universidad
de La Laguna, 0: 11-115.
Romaine, S. (1994). Language in Society. An Introduction to Sociolinguistics. Oxford: Oxford University Press.
Samper-Hernández, M. (2009). Evolución de la disponibilidad léxica en estudiantes grancanarios de enseñanza
primaria y secundaria. Las Palmas de Gran Canaria: Cabildo de Gran Canaria.
Samper, J. A. (2008). “Sociolinguistic aspects of Spanish in the Canary Islands”, The International Journal of the
Sociology of Language, 193/194: 161-176.
Samper, J. A. (dir.) (1998). Léxico del habla culta de Las Palmas de Gran Canaria. Las Palmas de Gran Canaria:
Cabildo Insular de Gran Canaria / Servicio de Publicaciones de la Universidad de Las Palmas de Gran
Canaria.
Trapero, M. and Llamas-Pombo, E. (1998). “Origen, etimología y significado de la palabra guanche”, Anuario de
Estudios Atlánticos, 44: 99-196.
Trapero, M. (1999). Pervivencia de la lengua guanche en el habla común de El Hierro. Santa Cruz de Tenerife:
Dirección General de Patrimonio Histórico. Gobierno de Canarias.
Trapero, M. (2007). Estudios sobre el guanche. La lengua de los primeros habitantes de las Islas Canarias. Las
Palmas de Gran Canaria: Fundación MAPFRE Guanarteme.
Trudgill, P. (1985). Sociolinguistics. An Introduction to Language and Society. Penguin.
Trujillo, R. (1981). “Algunas características de las hablas canarias”, Estudios colombinos, 2: 9-24.
Wardhaugh, R. (2002 [1986]). An Introduction to Sociolinguistics. Oxford, UK/Cambridge, USA: Blackwell.
Wölfel, D. J. (1996). Monumentae Linguae Canariae. Santa Cruz de Tenerife: Gobierno de Canarias. Dirección
General de Patrimonio Histórico, 2 vols. (translated into Spanish by Marcos Sarmiento-Pérez) (1st. ed. in
German, Graz, Austria, 1965).
RLyLA Vol. 09 (2014), 33-44 | 44
Vol. 9 año 2014, 45-54
EISSN 1886-6298
The role of derivational paradigms with adjectival base in Old English
word-formation
Carmen Novo Urraca
Abstract: The aim of this article is to identify the primary adjectives of Old English as well as to gather the derivational paradigms
that revolve around them. All in all, 459 primary adjectives are identified, which function as the base of 6,587 derivatives. Two
conclusions are drawn from the analysis. In the first place, the concept of derivational paradigm contributes to the explanation of
the overall organization of the lexicon, while allowing for the discussion of questions that are at the core of current morphological
theory, such as recursivity and productivity. Secondly, that primary adjectives play a significant role in Old English derivation. Even
though they are not as productive as strong verbs, primary adjectives function as base of derivation for a significant number of
non-basic terms, which, moreover, belong to all lexical categories and nearly all grammatical classes.
Keywords: Old English, word-formation, derivational paradigm, adjective, primary adjective, basic adjective.
1. INTRODUCTION
This article deals with Old English word-formation and, more specifically, with sets of derivatives that are
morphologically related to an adjective such as glēaw ‘penetrating, keen, prudent, wise, skilful; good’, as shown
below:*
(1)
a. Adjectives: ǣglēaw ‘learned in the law’, cræftglēaw ‘skilful, wise’, ferhðglēaw ‘wise, prudent’, foreglēaw
‘foreseeing, provident, wise, prudent’, frēaglēaw ‘very wise’, glēawferhð ‘prudent’, glēawhȳdig
‘thoughtful, wise, prudent’, glēawhycgende ‘thoughtful, wise, prudent’, glēawlic ‘wise, prudent, skilful,
diligent’, glēawmōd ‘wise, sagacious’, hreðerglēaw ‘wise, prudent’, hyrgeglēaw ‘prudent in mind’,
mōdglēaw ‘wise’, steorglēaw ‘clever at astronomy’, unglēaw ‘ignorant, foolish, unwise’, wordglēaw
‘skilful in words’.
b. Adverbs: foreglēawlīce ‘providently, prudently’, glēawe ‘wisely, prudently, well’ glēawlīce ‘prudently,
wisely, clearly, well’, unglēawlīce ‘unwisely, imprudently’.
c. Nouns: gereordglēawnes ‘skill in singing’, glēawnes ‘wisdom, prudence, skill, penetration; diligence;
sign, token’, glēawscipe ‘wisdom, thoughtfulness, diligence; proof, indication, test’, unglēawnes ‘folly,
ignorance’, unglēawscipe ‘folly’.
Example (1) shows members of the lexical categories adjective, adverb and noun resulting from morphological
processes of affixation and compounding on the base glēaw. In this study, a set of lexical items such as (1) is
termed a derivational paradigm, while the base of derivation of the paradigm is called the primary adjective. This
research aims to identify the primary adjectives in Old English as well as to gather the derivational paradigms that
revolve around them, which can be regarded as a contribution in a triple direction.
In the first place, this undertaking contributes to the refinement of the lexical data contained by the lexical
database of Old English Nerthus (www.nerthusproject.com). For this purpose, the whole database, including
approximately 30,000 entries, has been processed in order to gather the derivational paradigms of adjectives,
either by identifying completely new paradigms or by splitting the existing derivational paradigms of a strong verb.
| 45
Carmen Novo Urraca
The role of derivational paradigms with adjectival base in Old English word-formation
Secondly, this work contributes to the programme of research in the lexicology and derivational morphology
of Old English represented by Caballero González et al. (2004-2005), de la Cruz Cabanillas (2007), Ibáñez Moreno
(2007), González Torres (2009), Guarddon Anelo (2009a, 2009b), Martín Arista (2005, 2006, 2008, 2009, 2010a,
2010b, 2010c, 2011a, 2011b, 2011c, 2012a, 2012b, 2012c, 2013, fc.), Martín Arista and Cortés Rodríguez (fc.),
Martín Arista and Martín de la Rosa (2006), Pesquera Fernández (2009), Torre Alonso et al. (2008), Torre Alonso
(2009) and. This research continues the line of research opened by Kastovsky (1968, 1971, 1986, 1989, 1990,
1992, 2005, 2006). However, with the exception of two publications by Martín Arista (fc.-b, fc.-d), the lexical class
of the adjective has received little attention. For this reason, this journal article deals specifically with the adjective.
Regarding the question of derivational paradigms, Pesquera Fernández (2009) carried out a partial analysis of
strong verb paradigms. The methodology used in that study is adapted to apply to adjectival paradigms in the
present work.
Thirdly, this piece of research contributes to the compilation of the complete adjectival paradigms. In this
sense, it enlarges the analysis carried out by Seebold (1970), Heidermanns (1993) and Orel (2003), whose main aim
is the description of Germanic and, as a result, give only succinct information on Old English.
Previous work in the area of word-formation in the old Germanic languages has stressed the importance of
derivation from strong verbs, which has had two important effects on the proposals for the organization of the Old
English lexicon made so far. Firstly, that the importance of the process of zero derivation has been overestimated
and secondly, that the overall picture of the Old English lexicon is very conservative since it groups together many
lexical items that cannot be related to the base of derivation by means of a productive process. Consider, in this
respect, the following example:
(2)
grim ‘fierce, savage; dire, severe, bitter, painful’, ferhðgrim ‘savage’, grimmān ‘terrible sin’, heaðugrim
‘fierce’, heorugrimm ‘savage, fierce’, hetegrim ‘fierce, cruel’, hygegrim ‘savage, cruel’, nīðgrim ‘fierce,
hostile’, searogrim ‘fierce, formidable’, wælgrim ‘fierce, violent, bloody, cruel; fateful, dire’, āgrimsian
‘to provoke, irritate’, singrim ‘exceeding fierce’, unwælgrim ‘gentle, merciful’, grīmme ‘savagely, cruelly,
severely’, grimful ‘fierce, violent’, grimlīce ‘fiercely, severely, cruelly’, grimlic ‘fierce, blood-thirsty, cruel,
terrible, severe’, grimnes ‘ferocity, cruelty; severity’, grimsian ‘to rage’, grimsung ‘harshness, severity’,
wælgrimlīce ‘fiercely’, wælgrimnes ‘cruelty, torture’, grimena ‘caterpillar’, grimman ‘to rage; hasten on’.
If, following Hinderling (1967) and Kastovsky (1992), who hold that the strong verb is the starting point of
Germanic derivation, the strong verb grimman is considered the basic term from which all derivatives in the
paradigm derive, the adjective grim is zero derived from the strong verb and produces many compounds of its
own, including ferhðgrim, heaðugrim, heorugrimm, hetegrim, hygegrim, nīðgrim, searogrim, etc. If, on the contrary,
the adjective grim is the basic term around which the derivational paradigm is organized, this has the significant
consequence of reducing the scope of zero derivation to the formation of the strong verb grimman.
For the identification of the basic adjectives of Old English I draw on the study in Germanic primary adjectives
by Heidermanns (1993) and, to a lesser extent, on the etymological dictionaries by Seebold (1970) and Orel (2003).
As I have already pointed out, the data for the analysis have been retrieved from Nerthus (May 20, 2010), including
30,000 entries and the derivational paradigms of strong verbs.
2. THEORETICAL AND METHODOLOGICAL FOUNDATIONS
In a nutshell, some important aspects of previous research in the area on which I draw include: (i) the associative
character of the Old English lexicon, as described by Kastovsky (1992); (ii) the primary character of the Germanic
strong verb, which, as put forward by Hinderling (1976), Seebold (1970), Kastovsky (1992) and Bammesberger
(1992), represents the starting point of Germanic word-formation; (iii) the existence of a primary class of Germanic
adjectives which, at least partially, cannot be morphologically related to strong verbs, as Heidermanns (1993) claims;
(iv) the existence of a typological shift in the Old English lexicon, which changes from variable base morphology
to invariable base morphology, as Kastovsky (2006) puts it; and (v) the characteristics of recategorization and
recursivity as defining the derivational side of morphology, as Martín Arista (2009) has remarked.
I follow Pounder (2000) on the relevance of the notion of derivational paradigm, established as a counterpart
of the inflectional paradigm and with explanatory value at the paradigmatic dimension of the lexicon. For Pounder
(2000:82), the derivational paradigm is a set of paradigmatic relations between word-formations sharing a lexemic
root. This author distinguishes between the paradigm as a morphological structure, consisting of a set of paths
between a base and the operations that produce its derivatives, and the lexical paradigm involving a structured
RLyLA Vol. 09 (2014), 45-54 | 46
Carmen Novo Urraca
pattern of instructions for operations on stems. The former is valid for a whole lexical class and the latter is the
individual paradigm of a member of a lexical class.
The difference between the morphological and the lexical paradigm rests on the notions of degree of
abstractness and dynamism. The morphological paradigm defines a set of possible operations that are instantiated
(or not) by the lexical paradigm. Regarding this difference, Pounder (2000:86) excludes suppletion or insertion
from the paradigm of lexemes that do not hold a regular relation of form to the base of the paradigm because a
shared lexemic origin must be present for the establishment of a word-formation relation. Derivational paradigms
thus defined reinforce the associative character attributed to the Old English lexicon by Kastovsky (1992:294)
by considering series of derivatives like agan ‘to go’; began ‘to go over’, begang ‘practice’, beganga ‘inhabitant’,
begenge ‘practice’, bigengere ‘worker’, bigengestre ‘maiden’, etc. from gan ‘to go’. In Present-day English we find
a dissociated lexicon in which a Germanic lexical item can be related semantically to non-Germanic derivatives as
in mouth: oral, father: paternal, sun: solar, etc. As for the question of dynamism, the morphological paradigm, as a
set of operations, represents the dynamic part of word-formation, whereas the lexical paradigm, being a product,
constitutes the static part of word-formation.
Also of relevance to the distinction between the morphological and lexical paradigm is the question of
defectivity or formations resulting from correct paths (units and operations) in the morphological paradigm that
are not attested in the lexical paradigm. A case in point is provided by the compound glīwhlēoðriendlic ‘musical’,
the adjunct of which glīw is attested whereas its base hlēoðriendlic is not (Torre Alonso et al. 2008) although the
derivation of hlēoðriend from hleoðrian is stepwise, the adjective hlēoðriendlicø constitutes a lexical gap marked
with the subindex ø. The full derivation, consequently can be stated as hleoðrian > hleoðriend > hleoðriendlicø
> glīwhleoðriendlic the hypothetical predicate is an intermediate step between hleoðriend to glīwhleoðriendlic.
A derivation like hleoðrian > hleoðriend > hleoðriendlicø is well attested in other instances involving a present
participle of a weak verb to which the suffix -lic is attached, as in gītsiendlic ‘insatiable’, līciendlic ‘agreeable’,
mynegiendlic ‘hortatory’, sciriendlic ‘derivative’ and þrōwiendlic ‘suffering’.
On the question of zero derivation, I draw on González Torres (2009), who distinguishes between zero derivation
proper and derivation by inflectional means, and Martín Arista (2010a), who has put forward a typology of zero
derivation phenomena in Old English which includes: (i) zero derivation with explicit inflectional morphemes and
without explicit derivational morphemes; (ii) zero derivation without explicit or implicit morphemes, either inflectional
or derivational; (iii) zero derivation without inflectional or derivational morphemes but displaying ablaut; and (iv)
zero derivation with ablaut and formatives that can no longer be considered productive affixes, that is, umlaut and
inflection with derivational function (or at least, expressing morphological contrast) are included under the term of
zero derivation. This up-to-date revision of the notion of zero derivation has been central to the reorganization of a
number of adjectival paradigms analysed in this article.
As regards the notion of the adjective as the source of derivation, the crux of this research, I draw on Pilch
(1970), Campbell (1987), Kastovsky (1992) and Lass (1994). The lexical class of the adjective is the source of
derivation, by means of productive processes of affixation and compounding, of nouns, verbs, adverbs and other
adjectives. Beginning with nouns, deadjectival members of this category result from the attachment of the suffixes:
-dom, -ing, -ling, -ness, -scipe, -ð(o)/-t and –wist. The suffixes –cund, -fæst, -feald, -full, -ig, -lic, -sum and -wende
partake in the formation of deadjectival adjectives. Deadjectival verbs are formed by means of the attachment
of the following suffixes: -ettan, -læcan, -cian and –sian. It is important to note that the adjective is the main
source of derivation of the adverb. Nicolai (1907) distinguishes deadjectival adverbs in positive grade ending in
–e, deadjectival adverbs in positive grade ending in –līce, deadjectival adverbs in positive grade ending in –a,
and comparative grade adverbs and superlative grade adjectival adverbs. The adverbs derived from adjectives
by means of –e can be classified on thematic grounds as follows: a-stem, ja-stem, ja-stem with umlaut from the
adjective, wa-stem, and u-stem. The second group of Old English deadjectival adverbs, to which most Old English
adverbs belong, includes adverbs in positive grade ending in –līce. The classification is based on thematic vowel
of the adjective again: a-stem, ja-stem, wa-stem, u-stem.
There is a group of adverbs that take part in adjectival formations, including -sum, -bære, -fæst, -wende, -weard,
-cund and –feald. The adverbs in positive grade ending in –a represent the third subclass of deadjectival adverbs.
According to Campbell (1987), the suffix –a is characteristic of independent adverbs. The fourth subgroup of Old
English adverbs derived from adjectives consists of comparative grade adverbs, inflected for the comparative
or the superlative, both in –ost and –est. Turning to compounding, adjectives can be found in the formation of
nominal compounds in instances of adjective + noun and verbal adjective (past participle) + noun. In adjective
compounding, adjectives turn up in the following patterns: noun + adjective, adjective + adjective, adjective +
present participle, adjective + past participle and adverb + adjective.
RLyLA Vol. 09 (2014), 45-54 | 47
Carmen Novo Urraca
3. THE DERIVATIONAL PARADIGMS OF OLD ENGLISH PRIMARY ADJECTIVES
As has been remarked above, the aim of this article is to identify the primary adjectives of Old English as
well as to gather the derivational paradigms that can be built from them. The derivational paradigms that appear
below result from the processing of the whole lexical database of Old English Nerthus (www.nerthusproject.com),
which provides about 30,000 lexical entries as well as the derivational paradigms of strong verbs. With this basic
information, the compilation of the adjectival paradigms has entailed both the identification of completely new
adjectival paradigms and the splitting of the existing derivational paradigms of a strong verb into two parts, the
verbal part and the adjectival part. A more detailed account of the analysis process follows.
First of all, it has been necessary to classify adjectives on the grounds of their role in word-formation. There are
three possible types, namely derived adjectives, underived adjectives of the basic type and underived adjectives
of the primary type. For instance, ēstful ‘devout’ is a derived adjective because it results from the suffixation of the
affix -ful to the base of derivation ēste ‘gracious’. Underived adjectives of the basic type are adjectives that cannot
be related to a base of derivation by means of a morphological process of word-formation and, furthermore, do
not have derivatives of their own. For example, ǣdre ‘at once, directly, instantly, quickly; fully, entirely’ qualifies as
a basic adjective. The total figure of basic adjectives that have been found throughout the analysis is 118. Primary
adjectives constitute the base of derivation of derivational paradigms: 459 were identified in the analysis. As an
example the adjective dimm ‘dim, dark, gloomy, obscure; blurred, faint; wicked; wretched, grievous’ holds formal
and semantic relations to the following derivatives:
(3)
ādimmian ‘to become dim or dull, to darken, obscure’, dimhīw ‘of dark colour’, dimhof ‘place of
concealment’, dimhofe ‘place of concealment’, dimhūs ‘prison’, dimlic ‘dim, obscure, secret, hidden’,
dimmian ‘to be or become dim’, dimmnes ‘dimness, darkness, obscurity, gloom; evil, obscuration, moral,
obliquity; a dark place’, dimscūa ‘darkness, sin’, fordimmian ‘to obscure, darken’.
This paradigm is organised on the primary adjective dimm in such a way that all derivatives inherit formal and
semantic features from the primary adjective. The derivational paradigm of dimm comprises a total of 10 derivatives:
2 adjectives, 5 nouns and 3 verbs, which result from different morphological processes of word-formation. Primary
adjectives, therefore, generate a derivational paradigm consisting of all the words that are semantically and
morphologically related to the base of derivation, after which the paradigm is called. For instance, the primary
adjective gescēad ‘reasonable, prudent, calculated’ contains compounds such as gescādwīs ‘sagacious’ (from
gescēad ‘reasonable’ and wīs ‘wise’), zero derivatives like scēad ‘separation, distinction; understanding’, and affixal
derivatives of the type ungescēad ‘want of intelligence, senselessness’ and gescēadlīce ‘reasonably, rationally’.
The rest of the derivational paradigm of the primary adjective gescēad is given in (4) by morphological process:
(4)
a. Compounding: gescādwyrt ‘oxeye’, rihtgescēad ‘right understanding’, tungolgescēad ‘astrology’.
b. Prefixation: tōgescēadan ‘to expound, interpret’, ungescēadlic ‘unreasonable, indiscreet’,
ungescēadwīs ‘unintelligent, irrational, imprudent, foolish’, ungescēadlīce ‘unreasonably, absurdly’,
ungescēadwīslic ‘unintelligent, irrational, imprudent, foolish’.
c. Suffixation: (ge)scēadwīs ‘sagacious, intelligent, rational, reasonable, wise’, (ge)scēadwīslic
‘sagacious, intelligent, rational, reasonable, wise’, gescēadenlīce ‘severally’, gescēadlīce ‘reasonably,
rationally’, gescēadlic ‘reasonable, discreet, wise, accurate’, gescēadnes ‘a decree’, scēadwīsnes
‘sagacity, reason; discrimination, discretion; separation; reckoning’, ungescēadwīslīce ‘indiscreetly,
unreasonably, foolishly’, ungescēadwīsnes ‘want of intelligence, indiscretion, folly, ignorance’.
d. Zero derivation: ungescēad ‘unreasonable; excessive’.
As can be seen in the full derivational paradigm in (4), the question of recursivity arises. In effect, the gradual
derivation of ungescēadlic ‘unreasonable, indiscreet’ requires the previous derivation of ungescēad ‘unreasonable;
excessive’ from gescēad ‘reasonable, prudent; calculated’.
The second stage of the analysis has been to determine the category of the base of derivation of sets of
semantically and morphologically lexical items. In general, adjectives are the base of derivation of all categories,
except strong verbs (but see Martín Arista 2010a). New adjectival paradigms have been proposed whenever the
meaning of an adjective morphologically related to a strong verb differs from the meaning of the strong verb in
question. This is the case with dearf ‘bold’, and its derivatives dearflic ‘bold, presumptuous’, dearfscipe ‘boldness,
presumption’ and māndeorf ‘bold in evil?’, which are morphologically related to the strong verb (ge)deorfan ‘to
exert oneself, labour; be in peril, perish, be wrecked’. More problematic are instances in which adjectives that
are morphologically related to a strong verb do not have a different meaning to that of the strong verb. A relevant
example is provided by the pair calan ‘to grow cool or cold’ and ceald ‘cold’:
RLyLA Vol. 09 (2014), 45-54 | 48
Carmen Novo Urraca
(5)
ceald ‘cold’, ācealdian ‘to become cold’, ælceald ‘altogether cold, very cold’, brimceald ‘ocean-cold,
ceald ‘coldness, cold’, cealde ‘coldly’, cealdheort ‘cruel’, cealdian ‘to become cold’, cealdnes ‘coldness,
cold’, dægsceald ‘(shield by day?) sun’, hrīmceald ‘icy cold’, īsceald ‘icy cold’, oferceald ‘excessively
cold’, sincald ‘perpetually cold’, sincaldu ‘perpetual cold’, snāwceald ‘icy-cold’, wælceald ‘deadly-cold’,
winterceald ‘wintry-cold’.
As a general rule, a semantic criterion has been adopted regarding pairs like calan ‘to grow cool or cold’ and
ceald ‘cold’. Provided that the adjective cannot be decomposed semantically, it is considered the base of the
paradigm if the strong verb displays the adjective in its meaning definition and the adjective does not display the
strong verb similarly. Following this rule, the paradigm has been built on the base of ceald, rather than calan.
Thirdly, it has often been necessary to enlarge the inventory of adjectives provided by Heidermanns (1993). For
instance, this author lists getæl ‘numerous’, whereas, after retrieving the relevant information from Nerthus, the
resulting paradigm includes getæl 2 ‘numerous’, getæl 1 ‘number’ and talu ‘number’.
Fourthly, the overall organization of some paradigms as given by Nerthus has been modified. This is the case
with ēste ‘gracious, liberal’, given in (6):
(6)
ēste ‘gracious, liberal’, ēst ‘favour, grace, bounty, kindness, love; pleasure; harmony, consent’, ēstan ‘to
live luxuriously’, ēstelīce ‘courteously; luxuriously’, ēstelic ‘kind, gracious; devout; delicate, dainty (of ,
food)’, ēstful ‘gracious, devoted, devout; fond of’, ēstfullīce ‘kindly, devotedly’, ēstfulnes ‘devotion, zeal;
daintiness; luxury, lechery’, ēstgeorn ‘delicate, fond of luxuries’, ēstig ‘gracious, liberal’, ēstines ‘benignity’,
ēstmete ‘dainty (food), delicacy, luxury’, ēstnes ‘bliss’, ēstsum ‘freely, willingly, gladly’.
Finally, other paradigms have been newly created. This is the case with fals ‘false’, which includes fals
‘falsehood, fraud, counterfeit’. These instances give rise to a paradigm that has not been identified by previous
research. It comprises one derivative only, the zero derived noun fals.
By following the steps just described, the analysis of each derivational paradigm consists of the identification
of the primary adjective, the specification of derivatives by category and morphological process, the assessment
of recursivity and the information on etymology, as is demonstrated (7):
(7)
Primary adjective: æðele ‘noble, aristocratic, excellent, famous, glorious; splendid, fine, costly, valuable;
lusty, young; pleasant, sweet-smelling’.
Adjectives (9): æðelboren ‘of noble birth, distinguished; free-born; inborn, natural’, æðelcund ‘of noble
birth’, æðellic ‘noble, excellent’, efenæðele ‘equally noble’, fullæðele ‘very noble’, geæðele ‘natural,
congenial, suitable’, onæðele ‘natural to’, unæðelboren ‘low-born’, unæðele ‘of low birth, ignoble, base’.
Adverbs (2): æðellīce ‘nobly (BT)’, unæðelīce ‘ignobly, basely’.
Nouns (13): æðelborennes (f.) ‘nobility of birth or nature; inborn nature’, æðelcundnes (f.) ‘nobleness’,
æðelcyning (m.) ‘noble king’, æðelduguð (f.) ‘noble retinue’, æðeling (m.) ‘nobleman, chief, prince; king,
Christ, God; man, hero, saint; in pl. men, people’, æðelinghād (n.) ‘princely state’, æðelnes (f.) ‘nobility,
excellence’, æðelu (fn.) ‘nobility, family, descent, origin; nature; noble qualities, genius, talents, preeminence; produce, growth’, fæderæðelo (np.) ‘patrimony; paternal kinship’, rihtæðelcwēn (f.) ‘lawful wife’,
rihtæðelo (np.) ‘true nobility’, sibæðeling (m.) ‘related noble’, unæðelnes (f.) ‘ignobility’.
Weak verbs (2): geæðelian (2) ‘to make noble or renowned’, unæðelian (2) ‘to degrade, debase’.
Compounds (8): æðelboren ‘of noble birth, distinguished; free-born; inborn, natural’, æðelcyning (m.)
‘noble king’, æðelduguð (f.) ‘noble retinue’, efenæðele ‘equally noble’, fæderæðelo (np.) ‘patrimony;
paternal kinship’, rihtæðelcwēn (f.)
‘lawful wife’, rihtæðelo (np.) ‘true nobility’, sibæðeling (m.) ‘related noble’.
Prefixed (5): fullæðele ‘very noble’, onæðele ‘natural to’, unæðelboren ‘low-born’, unæðele ‘of low birth,
ignoble, base’, unæðelīce ‘ignobly, basely’.
Suffixed (9): æðelborennes (f.) ‘nobility of birth or nature; inborn nature’, æðelcund ‘of noble birth’,
æðelcundnes (f.) ‘nobleness’, æðeling (m.) ‘nobleman, chief, prince; king, Christ, God; man, hero, saint; in
pl. men, people’, æðelinghād (n.) ‘princely state’, æðellīce ‘nobly (BT)’, æðellic ‘noble, excellent’, æðelnes
(f.) ‘nobility, excellence’, unæðelnes (f.) ‘ignobility’.
Zero-derived (4): æðelu (fn.) ‘nobility, family, descent, origin; nature; noble qualities, genius, talents,
pre-eminence; produce, growth’, geæðele ‘natural, congenial, suitable’, geæðelian (2) ‘to make noble or
renowned’, unæðelian (2) ‘to degrade, debase’.
Recursive (aff>aff) (2): unæðelīce ‘ignobly, basely’, unæðelnes (f.) ‘ignobility’.
Etymology: aðal(j)a- (-ula-) ‘angestammt’; æðele, æðelra, æðerlest, geæðele, unæðele, æðelu, æðeling,
æðelic, geæðelian.
RLyLA Vol. 09 (2014), 45-54 | 49
Carmen Novo Urraca
The derivational paradigm in (7) comprises derivatives of the adverbial, adjectival, nominal and verbal classes,
among which nouns stand out, particularly affixed abstract nouns (æðelcundnes ‘nobleness’) and nominal
compounds with adjectival adjunct (æðelinghād ‘princely state’). Furthermore, derivatives are produced by all
major word-formation processes, including compounding, prefixation, suffixation and zero derivation. Affixation is
the most productive process, followed by compounding and zero derivation. Two instances of recursivity appear
in (7), unæðelīce ‘ignobly, basely’ (æðelīce) and unæðelnes ‘ignobility’ (æðelnes). The figure is not very high if
compared with the total derivatives of the paradigm, which displays 26 derived lexical items.
For the sake of comparison, the derivational paradigm of a primary adjective with fewer derivatives is offered
in (8):
(8)
Primary adjective: brant ‘deep, steep, high’.
Adjectives (1): brandstæfn ‘high-prowed?’.
Nouns (1): brenting (m.) ‘ship’.
Compounds (1): brandstæfn ‘high-prowed?’.
Zero-derived (1): brenting (m.) ‘ship’.
Etymology: branta- ‘hochragend’; brant (o), brenting.
In spite of the very low number of derivatives of the primary adjective brant ‘deep, steep, high’, it is noteworthy
that they fall into two lexical categories and are the result of two different word-formation processes. A discussion
of the results of the analysis illustrated by examples (7) and (8) follows in the next section.
4. DISCUSSION OF RESULTS
The adjective is relevant to word-formation in Old English for two reasons. On the quantitative side, a large
number of derivatives cluster around the paradigms of this lexical class. With a more qualitative approach,
deadjectival lexical items belong to all major lexical categories, including the noun, strong verb, weak verb, adjective
and adverb, as well as to other grammatical categories like the adposition, pronoun, numeral, conjunction and
interjection. I comment on these aspects in turn.
Beginning with the quantitative perspective, the analysis has demonstrated that the adjective is a very productive
lexical category. Only 118 basic (that is, without derivatives) adjectives have been identified, as opposed to the 459
primary adjectives that allow for the compilation of derivational paradigms. Moreover, 6,587 predicates have been
gathered around primary adjectives. This figure represents more than 20% of the Old English lexicon as listed by
Nerthus. Focusing on the lexical classes of the derivatives found in the paradigms, there are 2,821 nouns, 2,212
adjectives, 777 weak verbs, 673 adverbs and 72 strong verbs. In the grammatical classes we find 14 adpositions,
11 pronouns, 2 numerals, 2 interjections, 2 affixes and 1 conjunction. These data are shown in Table 1. The column
of derivatives displays the total number of instances found in the deadjectival paradigms analysed in this work.
The third column contains the percentage that the absolute figure constitutes within the total number of items of
the lexical category in question found in deadjectival paradigms. The fourth column presents the percentage of the
absolute figure in relation to the total number of items of the lexical category in the whole lexicon.
Table 1. Absolute and relative productivity by category.
Category
Derivatives
(absolute)
Relative to adjectival paradigms
(6,587=100%)
Relative to the whole lexical category
Nouns
2,821
42.82%
16.90 (16,689=100%)
Adjectives
2,212
33.58%
38.23 (5,785=100%)
Weak verbs
777
11.79%
19.71 (3,942=100%)
Adverbs
673
10.21%
40.68 (1,654=100%)
Strong verbs
72
1.09%
4.50 (1,598=100%)
Adpositions
14
0.21%
17.5 (80=100%)
Pronouns
11
0.16%
28.20 (39=100%)
Numerals
2
0.03%
3.84 (52=100%)
Interjections
2
0.03%
9.52 (21=100%)
Affixes
2
0.03%
1.17 (170=100%)
Conjunctions
1
0.01%
2.70 (37=100%)
RLyLA Vol. 09 (2014), 45-54 | 50
Carmen Novo Urraca
As can be seen in Table 1, nouns and adjectives together represent more than 75% of the derivatives of primary
adjectives. Weak verbs and adverbs form around 10% each, whereas the other categories are negligible on
quantitative grounds. Of special interest among these is the strong verb, with 1% relative to adjectival paradigms
and 4.5% relative to the whole lexicon. Table 1 also shows that approximately 40% of the lexical categories of
adverb and adjective come from the derivational paradigms of adjectives. The figure of nouns relative to the whole
lexical category is significantly lower (16%) than the figure relative to adjectival paradigms, in which 42.82% of
derivatives belong to this lexical category. The interpretation of these data is that deadjectival formation is not as
productive as other patterns for making new nouns. With respect to the whole lexicon, about 22% of the lexicon
belongs in the derivational paradigms of primary adjectives.
By process, there are 2,524 compounds and 2,034 suffixal derivatives. Suffixation is far more frequent than
prefixation, which contributes 997 derivatives. Zero derivation is practically as frequent as prefixation, giving
rise to a total of 928 zero derived predicates. Finally, conversion produces 101 derivatives. These figures are
displayed by Table 2, which is parallel to Table 1. The second column shows the total number of instances of each
morphological process found in deadjectival paradigms. The third column gives the percentage that the absolute
figure represents with respect to the total number of items produced by the process in question. The fourth column
presents the percentage represented by the absolute figure in relation to the total number of items produced by
the same process in the whole lexicon.
Table 2. Absolute and relative productivity by morphological process.
Processes
Derivatives
(absolute)
Relative to adjectival paradigms
(6,587=100%)
Relative to the whole process
Compounds
2,524
38.32
25.30 (9,975=100%)
Suffixed
2,034
30.88
30.79 (6,604=100%)
Prefixed
997
15.14
18.28 (5,454=100%)
Zero-derived
928
14.09
24.45 (3,794=100%)
Converted
101
1.53
46.33 (218=100%)
According to the data provided by Table 2, deadjectival derivational paradigms are mainly the product of
compounding and suffixation, these processes accounting for nearly 70% of the total amount of derivatives.
Prefixation and zero derivation produce around 15% each. It is noteworthy, in this respect, that suffixation has
twice as many derivatives as prefixation, which indicates a clear preference for the former, zero derivation also
representing a less productive process. When compared with the percentages found in the whole lexicon, it turns
out that affixation presents similar figures, whereas compounding is more frequent in deadjectival paradigms
than in the whole lexicon. On the other hand, zero derivation is less frequent in deadjectival paradigms than in the
lexicon. These figures show that adjectives resort to compounding (and secondarily to affixation) rather than zero
derivation for configuring their paradigms.
It is remarkable that recursive formations are not very frequent. Out of a total of 6,587 derivatives, only 413
are recursive. Recursive affixation (affix-affix) clearly outnumbers recursive compounding (compound as base
of further compounding). There are 342 recursive affixal derivatives as opposed to 71 instances of recursive
compounding only. It must be noted in this respect that I have analyzed only pure recursivity, that is, the feeding
of a morphological process by instances of the same process (Torre Alonso 2009), thus leaving aside cases
of affixation feeding compounding and compounding feeding affixation. In spite of this restriction, the figure of
recursive word-formation is low. Tentatively, it can be held that deadjectival derivatives are not very productive, but
more research is needed on this topic.
To conclude the quantitative analysis, a comparison with the paradigms of strong verbs is necessary. In the
latest report on the research carried out by the
Nerthus project, Martín Arista (2010c) has remarked that 359 derivational paradigms of strong verbs have
been gathered, in such a way that 18,850 predicates have been identified as belonging to these paradigms. This
makes an average type productivity of 52.5 derivatives per paradigm. There are 459 deadjectival paradigms, which
contain 6,587 derivatives, thus displaying an average productivity of type of 14.3 derivatives per paradigm. There
are reasons to hold, consequently, that the derivational paradigms of adjectives are not as productive as those of
strong verbs.
RLyLA Vol. 09 (2014), 45-54 | 51
Carmen Novo Urraca
On the qualitative side, this research has demonstrated that, along with the strong verb, the adjective plays
a significant role in the organization of the Old English lexicon. This agrees with Heidermanns (1993), who has
identified the primary adjectives of Germanic, rather than Bammerberger (1965), Hinderling (1967), Seebold (1970)
and Kastovsky (1968, 1992), since the latter authors have stressed the central character of the strong verb to
Germanic derivation, thus paying little attention to the derivatives based on other lexical categories. Apart from the
quantitative data offered above, it must be noted that the only two categories that are not found within adjectival
paradigms are the demonstratives/articles and the possessives. Also of qualitative relevance is the fact that
adjectives constitute bases of derivation for strong verbs. Even though the question has been tackled by previous
research, thus Pilch (1970) and Martín Arista (2010b), this is the first full report on the deadjectival Old English
strong verbs. The list includes derivatives that belong to all seven classes of strong verbs:
(9)
Class I (14): ætslīdan ‘to slip, glide, fall’, āglǣdan ‘to cause to slip’, āglīdan ‘to glide, slip’, āslīdan ‘to slide,
slip, fall’, efenwrītan ‘conscribere, to set down, present’, forðbesēon ‘to look forth’, forðgewītan ‘to go
forth, pass; depart, die’, (ge)bītan ‘to bite, tear; cut’, glǣdan ‘to cause to slip or totter’, glīdan ‘to glide, slip,
slide’, niðerāstīgan ‘to descend’, niðerstīgan ‘to descend’, tōglīdan ‘to glide away, split, slip’, ūtāslīdan ‘to
slip forwards, fall (into)’.
Class II (12): ālēogan ‘to lie, deny, deceive’, belēosan ‘to be deprived of, lose’, forlēosan ‘to lose, abandon’,
forðgēotan ‘to pour forth’, forð hrēosan ‘to rush forth’, forð tēon ‘to draw forth, bring forth’, fullflēon ‘to
take to flight, escape’, (ge)lēogan ‘to lie; deceive’, nið erāscūfan ‘to push down’, oflēogan ‘to lie, be false’,
rēodan ‘to redden, stain with blood’, sēocan ‘to be ill, fall ill’.
Class IIIa (3): āswindan ‘to become weak, shrink’, grimman ‘to rage; hasten on’, wiðerwinnan ‘to revolt’.
Class IIIb (5): forðyrnan ‘to run before, precede’, (ge)sceorpan ‘to scrape, gnaw’, mæltan ‘to consume by
fire, melt, burn up’, sciellan ‘to sound (Sweet)’, smeortan ‘to smart’.
Class IIIc (1): fullberstan ‘to burst completely’.
Class IV (4): efencuman ‘to come together, agree’, forðbecuman ‘to come forth’, forðberan ‘to bring forth,
produce’, forðcuman ‘to come forth, proceed, arrive at, succeed’.
Class V (12): æftercweðan ‘to speak after, repeat, answer’, efenāmetan ‘to assemble together; compare’,
efenetan ‘to eat as much as’, efengefēon ‘to rejoice together’, efengemetan ‘to compare’, efenmetan ‘to
assemble together; compare’, forðwegan ‘to further; to advance’, hearmcweðan ‘to speak evil of’,
repan ‘to reap’, wið ercweð an
‘to withstand’, wið ermetan ‘to compare’,
wyrgcweðan.
Class VI (3): forðstæppan ‘to issue forth, proceed’, mānswerian ‘to forswear, perjure oneself’, wiðerstandan
‘to resist’.
Class VII (a) (1): tōgescēadan ‘to expound, interpret’.
Class VII (c) (6): efenweaxan ‘to grow together’, forð weaxan ‘to break forth, burst forth’, fullweaxan ‘to
grow to maturity’, (ge)manigfealdan ‘to multiply, abound, increase’, (ge)wealcan ‘to move round, revolve,
roll, toss’, onwealcan ‘to roll, roll round’.
Class VII (d) (1): forðgangan ‘to go forth’.
Class VII (e) (4): forebrǣdan ‘to prolongue; overshadow’, (ge)brǣdan ‘to make broad, extend, spread’,
gelīclǣ tan ‘to liken, compare’, oferbrǣ dan ‘to spread over, suffuse’.
Class VII (f) (2): æfterrōwan ‘to row after’, fullgrōwan ‘to grow to perfection’.
Strong with weak forms (4): (ge)munan ‘to think about, remember’, gemynan ‘to think about, be mindful of’,
ofmunan ‘to remember, collect’, onmunan ‘to esteem, care for, wish’.
Finally, in relation to the qualitative approach, it must be stressed that the derivational paradigms of adjectives
have allowed us to identify two affixes that have undergone a process of grammaticalization, thus turning into a
derivational affix from a lexical category, namely -sām and -healf.
5. CONCLUDING REMARKS
Two main conclusions can be drawn from this research. In the first place, the concept of derivational paradigm
contributes to the explanation of the overall organization of the
lexicon, while allowing for the discussion of questions that are at the core of current morphological theory, such
as recursivity and productivity. Secondly, primary adjectives play a significant role in Old English derivation. Even
though they are not as productive as strong verbs, it cannot be denied that they are the base of derivation of a
significant number of non-basic terms, which, moreover, belong to all lexical categories and nearly all grammatical
categories.
RLyLA Vol. 09 (2014), 45-54 | 52
Carmen Novo Urraca
NOTES
This research has been funded through the project FFI2011 29532. The following abbreviations are used throughout the
article: ADJ (adjective), ADV (adverb), BT (Bosworth Toller), f. (feminine), m. (masculine), n. (neuter), N (noun) and p. (plural).
REFERENCES
Bammesberger, A. (1965). Deverbative jan-Verba des Altenglischen, vergleichend mit den übrigen altgermanischen
Sprachen dargestellt. München: Ludwig-Maximilians U.
Bammesberger, A. (1992). “The place of English in Germanic and Indo-European”, R. Hogg (ed.) The Cambridge
History of the English Language I: The Beginnings to 1066. Cambridge: Cambridge UP. pp. 26-66.
Caballero González, L., González Torres, E., Ibáñez Moreno, A. and Martín Arista, J. (2004-2005). “Predicados
verbales primitivos y derivados en inglés antiguo. Implicaciones para la elaboración de una base de datos
lexica”. Revista Española de Lingüística Aplicada, 17-18: 35-49.
Campbell, A. (1987 [1959]). Old English Grammar. Oxford: Oxford UP.
de la Cruz Cabanillas, I. (2007). “Semantic primes in Old English: a preliminary study of descriptors”. SELIM, 14:
37-58.
González Torres, E. (2009). Affixal Nouns in Old English: Morphological Description, Multiple Bases and Recursivity.
Unpublished Ph. D. thesis. Universidad de La Rioja, Spain.
Guarddon Anelo, M. C. (2009a). “The Natural Semantic Metalanguage of Old English Compound Adpositions”,
ES-English Studies, 30: 61-83.
Guarddon Anelo, M. C. (2009a). “Un análisis de las propiedades combinatorias de los primitivos semánticos a
través de las adposiciones complejas en inglés antiguo”. Revista Española de Lingüística. Forthcoming.
Heidermanns, F. (1993). Etymologisches Wörterbuch der germanischen Primäradjektive. Berlin: Walter de Gruyter.
doi:10.1515/9783110871616
Hinderling, R. (1967). Studien zu den starken Verbalabstrakten des Germanischen. Berlin: Walter de Gruyter.
doi:10.1515/9783110822663
Ibáñez Moreno, A. (2007). A Functional Approach to the Derivational Morphology of Old English: Verbal Suffixes.
Unpublished Ph. D. thesis. Universidad de La Rioja, Spain.
Kastovsky, D. (1968). Old English Deverbal Substantives Derived by Means of a Zero Morpheme. PhD Dissertation.
Tübingen: Eberhard-Karls-U.
Kastovsky, D. (1971). “The Old English Suffix -ER(E)”. Anglia LXXXIX, 3: 285-325.
Kastovsky, D. (1986). “Deverbal nouns in Old and Modern English: from stem-formation to word-formation”, J.
Fisiak (ed.) Historical Semantics-Historical Word Formation. Berlin: Mouton de Gruyter. pp. 221-61.
Kastovsky, D. (1989). “Typological Changes in the History of English Morphology”. U. Fries and M. Heusser (eds.)
Meaning and Beyond. Ernst Leisi zum 70. Geburstag. Tübingen. pp. 281-93.
Kastovsky, D. (1990). “The typological status of Old English Word Formation”, S. M. Adamson, V. A. Law, N.
Vincent and S. Wright (eds.) Papers from the 5th International Conference on English Historical Linguistics.
Amsterdam: John Benjamins. pp. 205-24. doi:10.1075/cilt.65.14kas
Kastovsky, D. (1992). “Semantics and vocabulary”. R. Hogg (ed.) The Cambridge History of the English Language
I: The Beginnings to 1066. Cambridge: Cambridge UP. pp. 290-408.
Kastovsky, D. (2005). “Conversion and/on zero: word-formation theory, historical linguistics, and typology”, in L.
Bauer and S. Varela (eds.) Approaches to Conversion/Zero Derivation. Münster: Waxmann. pp. 31-50.
Kastovsky, D. (2006). “Typological Changes in Derivational Morphology”. A. van Kemenade and B. Los (eds.) The
Handbook of The History of English. Oxford: Blackwell, 151-77.
Martín Arista, J. (2005). ‘Ge- and the descriptive power of Nerthus’. Journal of English Studies, 5-6: 209-31.
Martín Arista, J. (2006). “Alternations, relatedness and motivation: Old English A-”, in P. Guerrero Medina and E.
Martínez Jurado (eds.) Where Grammar Meets Discourse Functional and Cognitive Perspectives. Córdoba:
Servicio de Publicaciones de la U. de Córdoba. pp. 113-32.
Martín Arista, J. (2008). “Unification and separation in a functional theory of morphology”, in R. Van Valin (ed.)
Investigations of the Syntax-Semantics-Pragmatics Interface. Amsterdam: John Benjamins. pp. 119-45.
doi:10.1075/slcs.105.12mar
Martín Arista, J. (2009). “A Typology of Morphological Constructions”, in C. Butler and J. Martín Arista (eds.)
Deconstructing Constructions. Amsterdam: John Benjamins. pp. 85-115. doi:10.1075/slcs.107.06aty
Martín Arista, J. (2010a). “OE strong verbs derived from strong verbs”, SKASE Journal of Theoretical Linguistics,
7-1: 36-56.
RLyLA Vol. 09 (2014), 45-54 | 53
Carmen Novo Urraca
Martín Arista, J. (2010b). “Lexical negation in Old English”, NOWELE-North-Western European Language Evolution
60/61: 89-108
Martín Arista, J. (2010c). “Building a lexical database of Old English: issues and landmarks”, in J. Considine (ed.)
Current projects in historical lexicography. Newcastle: Cambridge Scholars Publishing. pp. 1-36.
Martín Arista, J. (2011a). “Projections and Constructions in Functional Morphology. The Case of Old English
HRĒOW”, Language and Linguistics 12/2: 393-425.
Martín Arista, J. (2011b). “Adjective formation and lexical layers in Old English”, English Studies 92/3: 323-344.
Martín Arista, J. (2011c). “Morphological relatedness and zero alternation in Old English”, in P. Guerrero Medina
(ed.), Morphosyntactic Alternations in English. Sheffield/Oakville: Equinox. pp. 339-362.
Martín Arista, J. (2012a). “Morfología flexiva en RRG”, in R. Mairal, L. Guerrero and C. González (eds.), El
funcionalismo en la teoría lingüística. Madrid: Akal Universitaria. pp. 43-58.
Martín Arista, J. (2012b). “The Old English Prefix Ge-: A Panchronic Reappraisal”, Australian Journal of Linguistics
32/4: 411-433. doi:10.1080/07268602.2012.744264
Martín Arista, J. (2012c). “Lexical database, derivational map and 3D representation”, RESLA Revista Española de
Lingüística Aplicada, 26/Extra issue: 119-144.
Martín Arista, J. (2013). “Recursivity, derivational depth and the search for Old English lexical primes”, Studia
Neophilologica 85/1: 1-21. doi:10.1080/00393274.2013.771829
Martín Arista, J. Noun layers in Old English. Forthcoming.
Martín Arista, J. and F. Cortés Rodríguez. “From directionals to telics: meaning construction, word-formation
and grammaticalization in Role and Reference Grammar” in M. A. Gómez González, F. Ruiz de Mendoza
Ibáñez and F. Gonzálvez García (eds.), Form and function in language: Functional, cognitive and applied
perspectives. Amsterdam: John Benjamins. Forthcoming.
Martín Arista, J. and Martín de la Rosa, V. (2006). “Old English Semantic Primes: Substantives, Determiners and
Quantifiers”, ATLANTIS 17: 9-28.
Nicolai, O. (1907). Die Bildung des Adverbs im Altenglischen. Kiel: H. Fiencke.
Orel, V. (2003). A Handbook of Germanic Etymology. Leiden: Brill.
Pesquera Fernández, L. (2009). Transparent and Opaque Word-Formation in the Derivational Paradigm of Old
English Strong Verbs. Unpublished Ph.D. thesis. Universidad de La Rioja, Spain.
Pilch, H. (1970). Altenglische Grammatik. München: Max Hueber Verlag.
Pounder, A. (2000). Processes and Paradigms in Word-Formation Morphology. Berlin: Mouton de Gruyter.
doi:10.1515/9783110814378
Seebold, E. (1970). Vergleichendes und etymologisches Wörterbuch der germanischen starken Verben. The Hague:
Mouton. doi:10.1515/9783110821956
Torre Alonso, R., Martín Arista, J., Ibáñez Moreno, A., González Torres, E. and Caballero González, L. (2008).
“Fundamentos empíricos y metodológicos de una base de datos léxica de la morfología derivativa del
inglés antiguo”, Revista de Lingüística y Lenguas Aplicadas 3: 129-44. doi:10.4995/rlyla.2008.696
RLyLA Vol. 09 (2014), 45-54 | 54
Vol. 9 año 2014, 55-66
EISSN 1886-6298
Alternations vs. variation in Old English. Methodological and
descriptive issues
Abstract: This paper addresses the phenomena of variation and alternation as reflected by the Old English version of
De Temporibus Anni (Blake 2009). The analysis, which focuses on the stem vowels of verbs and is based on the diatopic and
diachronic contrasts found by de la Cruz (1986) as well as the alternations identified by Kastovsky (1968), aims at deciding if a
given equivalence is a product of variation or alternation. The results indicate that alternation is a more predictable and systematic
phenomenon than variation. The conclusions stress the importance of the diphthong <ie> for distinguishing between Early and
Late West-Saxon and as a point of contact between the phenomena of i-mutation, alternation and diachronic variation.
Keywords: Old English, alternation, dialectal variation, diachronic variation.
1. Introducción
The aim of this article is to draw a distinction between the concepts of alternation and variation in Old English.1
These concepts are illustrated and discussed with reference to the Old English version of De Temporibus Anni (Blake
2009). The focus of the analysis is on vocalic alternations and variation in verbs. Throughout the discussion, some
methodological and descriptive issues are raised that allow to reach the conclusion that variation is a less systematic
phenomenon than alternation. Thus presented, this article can be seen as a contribution to the study of Old English
morphology, which has pursued two main lines of research. In the first place, Kastovsky (1986, 1989, 1990, 1992,
2006) identifies a typological change in the morphology of Old English as a result of which invariable bases replace
variable bases of inflection and derivation. According to Kastovsky (2006), this evolution takes place in two steps:
from root-formation to stem-formation and, later on, from stem-formation to word-formation. In the same line,
Haselow (2011) finds some analytic tendencies in the derivational morphology of English that originate in the change
from variable to invariable base morphology. Secondly, Martín Arista (2008, 2009, 2010a, 2010b, 2011a, 2011b,
2011c, 2012a, 2012b, 2013, fc.-a, fc.-b) develops a functional theory of morphology focusing on some aspects of
functional grammars like layering and projection and applies it to the inflection and derivation of Old English, as a
result of which some lexical layers can be defined on the grounds of different morphological processes and different
degrees of morphological productivity.
The outline of the article is as follows. Section 2 presents the framework of alternations, while section 3
presents the relevant patterns of variation of the diachronic and dialectal types. Section 4 discusses the instances
of alternations and variation that arise in De Temporibus Anni and, to close this article, section 5 draws the main
conclusions.
2. THE FRAMEWORK OF ALTERNATIONS
Ray (1996:13) remarks that the origin of Germanic strong verbs is to be found in primary Indo-European verbs,
which, unlike other verbs like denominatives, causatives and iteratives, were not derived. Whereas Germanic
strong verbs developed from primary verbs, weak verbs developed from derived Indo-European verbs. Rix et al.
1
This research has been funded through the project FFI2011-29532.
| 55
Alternations vs. variation in Old English. Methodological and descriptive issues
(2001) provide the Indo-European reconstruction of Germanic strong verbs, which, for the letter B, includes the
following verbal lexemes:
(1) bacan ‘to bake’ < *bheh3g-, bēatan ‘to beat’ < bheud-, belgan ‘to be angry’ < *bhelgh-, bellan ‘to bellow’
< *bhelH-, bendan ‘to bend’ < *bhendh-, blīcan ‘to glitter’ < *bhleig-, blōwan ‘to blow’ < *bhleh2g-, brēowan
‘to brew’ < *bhreuH-, brūcan ‘to brook’ < *bhreuHg-, būgan ‘to bow’ < *bheugh-.
Indo-European primary verbs can be divided into non-thematic, which did not insert a thematic vowel after the
root, and thematic, which had stems ending in thematic vowels (e, o). Szemerényi (1996) notes that non-thematic
verbs disappeared in Germanic and reduplication, which had been a central formative principle of the perfect stem
in Indo-European, was dropped. As a result, gradation or ablaut becomes the only formative principle available
after the Germanic simplification of the Indo-European verb system. In Ray´s (1996:68) words:
[gradation] became the chief means of expressing the temporal relations of the verb. This is not surprising, for
the principles of ablaut (…) ran through the whole language and (...) it was rare to find vowels (…) playing part in
important linguistic processes (…) which were not subject to it. (Ray 1996:68)
For Ringe (2006:10), the Indo-European patterns of alternation ē ~ e ~ Ø ~ o ~ ō and ā ~ a ~ Ø have the
following reflexes in Germanic:
(2) Short series
e ~ Ø ~ o
a ~ Ø ~ o
o ~ Ø ~ o
Long series
ē~a~ō
ā~a~ō
ō~a~ō
The Proto-Germanic strong verbs display the patterns in (2) as shown by (3), where C stands for consonant, R
for resonant and N for nasal (Ringe 2006: 11):
(3) Class I
Class II
Class III
Class IV
Class V
Class VI
īC ~ aiC ~ iC
euC ~ auC ~ uC
eRC/iRC ~ aNC ~ uNC
eR/iR ~ aR ~ uR
eC/iC ~ aC ~ēC
aC ~ ōC ~ ōC
As Ray (1996:97) points out, the correspondence between Proto-Germanic and Germanic strong verbs can be
accounted for in the following terms. Germanic strong verbs can be classified into e-verbs, a-verbs, ǣ-verbs and
ō-verbs, in such a way that e represents Proto-Germanic strong grade e, a corresponds to strong grade o or a, ǣ
stands for strong grade ē and ō is the reflex of strong grade ō or ā. Such vocalic correspondences are reflected
in the seven classes of strong verbs found in Old English and the other old Germanic languages. The basis of the
classification is the vowel alternations holding among the infinitive, the preterite singular, the preterite plural and
the past participle, as presented in (4):
(4)
Infinitive
Preterite singular
Preterite plural Past Participle
Idrifan ‘to drive’drāfdrifondrifen
IIclēofan ‘to cleave’clēafclufonclofen
IIIdrincan ‘to drink’drancdruncondruncen
IVberan ‘to bear’bærbǣron
geboren
Vgifan ‘to give’geafgēafongifen
VIstandan ‘to stand’stōdstōdonstanden
VII slǣpan ‘to sleep’slēpslēponslǣpen
As Hogg and Fulk (2011) explain, the infinitive of class I has ī followed by one consonant, as in scīnan ‘to shine’.
The infinitive of class II has either ēo or ū followed by one consonant as in, respectively, crēopan ‘to creep’ and
brūcan ‘to enjoy’. The infinitives of class III can be grouped under five sub-classes: e followed by two consonants
(bregdan ‘to move’), eo + r/h plus consonant (beorgan ‘to protect’), l plus consonant (helpan ‘to help’), palatal
plus ie followed by two consonants (gieldan ‘to pay’) and i + nasal followed by consonant (drincan ‘to drink’). The
infinitive of class IV has e plus liquid, as in beran ‘to bear’. The infinitive of class V has e plus plosive or fricative,
as in cweðan ‘to say’. Finally, the distinctive characteristic of class VII is that it has the same vowel either in
the infinitive and the past participle (dragan ‘to draw’, past participle dragen) or in the preterite singular and the
preterite plural (hēold, hēoldon, preterite of healdan ‘to hold’).
RLyLA Vol. 09 (2014), 55-66 | 56
Hinderling (1967) and Kastovsky (1992, 2006) consider the strong verb the starting point of lexical derivation in
Germanic. As can be seen in (5), all the major lexical categories (including the strong verb itself) have derivatives
based on strong verbs in Old English:
(5)a.
Strong verb-noun (Palmgren 1904; Kastovsky 1968)
cweðan ‘to say’ ~ cwiss ‘speech’
fōn ‘to take’ ~ feng ‘grasp’
hweorfan ‘to turn’ ~ hwearft ‘revolution’
b.
Strong verb-weak verb (Schuldt 1905; Bammesberger 1965)
hwīnan ‘to hiss’ ~ āhwǣnan ‘to afflict’
belīfan ‘to remain over’ ~ lǣfan ‘to leave’
scīnan ‘to shine’ ~ gescǣnan ‘to break’
c.
Strong verb-adjective (Schön 1905)
biddan ‘to ask’ ~ bedul ‘suppliant’
magan ‘to be able’ ~ meagol ‘mighty, strong’
stīgan ‘to move’ ~ stǣgel ‘steep’
d.
Strong verb-strong verb (Martín Arista 2010a, 2010b)
faran ‘to go, fare’: āfaran ‘to go out’, befaran ‘to go’, forfaran ‘to pass away’, forðfaran ‘to depart’,
forðgefaran ‘to depart’, infaran ‘to enter’, misfaran ‘to go wrong’, offaran ‘to interpret’, oferfaran
‘to pass’, oðfaran ‘to free from’, tōfaran ‘to be scattered’, ðurhfaran ‘to pass through’, ūtfaran ‘to
go out’, ūtāfaran ‘to come forth’, wiðfaran ‘to come off’, ymbfaran ‘to surround’.
The derivation of nouns from verbs has drawn more attention than other lexical categories. Palmgren (1904)
classifies strong verb derivatives on the grounds on the part of the verb, that is, present tense formations, past
formations and past participle formations. Present tense formations include neuters (hlæd ‘mound, pile’~ hladan
‘to lade’), strong masculines (hrōp ‘clamour’~ hrōpan ‘to shout’), weak masculines (hweorfa ‘whorl of spindle’ ~
hweorfan ‘turn’), strong feminines (span ‘span’ ~ spannan ‘clasp’) and weak feminines (dræge ‘drag-net’~ dragan
‘to drag’). Past tense formations can be broken down into neuters (mealt ‘malt’~ meltan ‘to consume by fire’),
strong masculines (including o- stems like scēaf ‘bundle, sheaf’~ scūfan ‘to shove’; and i- stems like hlīet ‘lot’~
hlēotan ‘to cast lots’), weak masculines (wrēcca ‘exile, wretch’~ wrecan ‘to drive’), feminines based on the vowel
of the singular (including strong feminines with short stem-syllable like cwalu ‘killing’ ~ cwelan ‘to kill’; strong
feminines with long stem-syllable like bād ‘pledge, impost; expectation’~ bīdan ‘to stay’; and weak feminines
like scīete ‘sheet, cloth’~ scēat ‘napkin’. Past tense formations based on the vowel of the plural comprise strong
feminines (scēara ‘shears’~ scieran ‘to cleave’) and weak feminines (wǣge ‘scales, balance’~ wegan ‘to weigh’).
Past participle formations can be of the neuter gender (swol ‘burning’~ swelan ‘to burn’); of the masculine gender,
including strong nouns (with short stem-syllable, such as bryne ‘burning’~ biernan ‘to burn’; and long stem-syllable
such as swēg ‘noise, sound’~ swōgan ‘to resound’) and weak nouns (unna ‘permission’~ unnan ‘to grant’); and
the feminine gender (both strong like hulu ‘husk’~ helan ‘to calumniate’; and weak like storfe ‘flesh of animals that
have died by disease’~ steorfan ‘to die’).
The instances of strong verb-noun derivation presented above show that, while some formations keep the
vocalic grade of the strong verb base, thus hrōp ~ hrōpan, it is usually the case that there is a contrast between
the vocalic grade of derived noun and the strong verb, as in cwiss ~ cweðan, feng ~ fōn and hwearft ~ hweorfan.
Moreover, these vocalic contrasts tend to be recurrent, as can be seen in the following derivatives of the strong
class III, all of which have a front vowel /e/ as opposed to the back vowel /a/ of the preterite form of the strong
verb, on which they are based:
(6)
hlēmm ‘sound, noise’ ~ hlimman ‘to resound, roar’
stēng ‘pole, stake’ ~ stingan ‘to sting’
swēnc ‘tribulation’ ~ swincan ‘to toil’
The vocalic contrasts arising in (6) have been largely discussed in the literature as ablaut (or apophony) and the
different vocalic values are usually referred to as ablaut grades. From the morphological point of view, recurrent
contrasts between related forms that share a lexemic root have been dealt with in terms of alternations. Vocalic
alternations in Old English are motivated by i-mutation, which is described by Hogg (1992:113) as follows: Old
English vowels harmonised to an /i/ or /j/ following them in the same word. This caused all back vowels to front
and all short vowels (except naturally /i/) and diphthongs to rise when /i/ or /j/ followed in the next syllable. The
fronting of the back vowels /o/ and /u/ went through intermediate rounding, so that /o/ > /oe/ > /e/ and /u/ > /ue/
(<y>) > /i/. In general, previous research concurs that the phonological rules that produced ablaut were eventually
morphologized (thus Lass 1994:108; Ringe 2006:10). In Kastovsky´s (1968:58) words:
RLyLA Vol. 09 (2014), 55-66 | 57
The alternations are in general only historically motivated, the motivation having disappeared on account of various
sound changes long before the beginning of the literary period (...) The originally phonologically conditioned
alternations thus became non-automatic, unpredictable, and must therefore be considered morphologically
conditioned in OE, which permits us to conclude that they probably were no longer productive. (Kastovsky 1968:58)
Kastovsky (2006:171) dates the disappearance of alternations in the Middle English period, although he
remarks that at the end of the Old Enlish period these alternations were already unpredictable. In spite of their
morphological conditioning, Old English alternations hold between the strong verb on the one hand and the noun,
the adjective and the weak verb on the other.
Figure 1 illustrates the alternations obtaining between nouns and strong verbs and figure 2 the alternations
holding between nouns and weak verbs. In Kastovsky´s (1968) analysis of Old English alternations, direct
alternations conform to i-mutation whereas reverse alternations do not. In figures 1 and 2, A indicates that the
alternation is vocalic, while R marks a reverse alternation.
DirectNoun
A1
fær
A2
ece
A3
fiell
A4a
cwide
A4b
wierp
A5
bryce
A6
byrst
A7
drǣf
A8
swēg
A9
hlīet
A10
flīeta
A11
Verb
Reverse Noun Verb
faranA1Rstalu
stæl (stelan)
acanA2Rswaruswerian
feallanA3R
cweðen (cweðan)
A4aR
weorpanA4bR
brocen (brecan)
A5R
burston (berstan)
A6R
drāf (drīfan) A7R
swōganA8R
hlēat (hlēotan)
A9R
flēotanA10R
Figure 1: Strong verb vocalic alternations.
DirectNoun
Verb
Reverse Noun
A1
A2
cempa
campiansand
A3A3Rgemearr
A4a
A4aR
A4b
fyrm
feormianA4bR
weorc
A5
A5R
spor
A6
A6R
husc
A7
ǣsce
āscianA7Rlār
A8
lēc lōcian A8Rfōda
A9
A9R
ēaca
A10
frīg frēoganA10Rstēora
A11
bȳ būan
Verb
sendan
mierran
wyrcan
spyrian
hyscan
lǣran
fēdan
īecan
stīeran
Figure 2: Weak verb vocalic alternations.
Figure 1 describes a situation in which the derivation of nouns from strong verbs mainly produces direct
alternations, as in ece ‘pain’ ~ acan ‘to ache’, where the noun displays the predictable ablaut grade with respect
to the verb on the basis of i-mutation. Figure 2, on the other hand, mainly contains reverse alternations, in such
a way that the direction of i-mutation in pairs like fōda ‘food’ ~ fēdan ‘to feed’ clearly indicates that the derivation
must have started in the noun, the weak verb representing the i-mutation of the noun. Two remarks must be made
on this interpretation of Kastovksy´s (1968) alternations. The first has to do with the terms direct and reverse. The
direction of i-mutation is constant. Both in weorpan ~ wierp and būan ~ bȳ a front vowel and a back vowel are
related to each other by i-mutation. Considering the base category, however, it is the case that the noun wierp
is the i-mutation of the strong verb weorpan while the weak verb būan represents the i-mutation of the noun bȳ.
In other words, the category rather than the vocalic grade is reversed. Whereas the strong verb is the base of
derivation with respect to the noun, the noun is basic with respect to the weak verb. This is in keeping with the
status of starting point of lexical derivation enjoyed by the strong verb, but has two important consequences.
Firstly, if cempa derives from campian it follows that not all weak verbs are derived. Secondly, if swerian is the
base of derivation of swaru not all strong verbs are basic. With these premises, it must be noted that the reverse
alternations A4aR and A11R are unattested in Kastovsky´s (1968) account.
RLyLA Vol. 09 (2014), 55-66 | 58
3. THE FRAMEWORK OF VARIATION
This section presents the patterns diatopic (interdialectal) and diachronic (intradialectal) variation as rendered in
previous research, mainly by de la Cruz (1986). The focus is on vocalic variation. In general, interdialectal contrasts
are described with reference to West-Saxon while intradialectal contrasts refer to West-Saxon exclusively.
Beginning with interdialectal vocalic variation, the contrast <æ>/<e> distinguishes the West-Saxon dialect
from the other Old English varieties. West-Saxon prefers the forms in <æ>, while <e> varieties are used in other
dialects. This is the case with the preterite of the verb beran ‘to bring’, which is bær in West-Saxon but ber in
Kentish and Southern Mercian; the infinitive lǣtan ‘to leave’ and sǣtan, the preterite form of sittan ‘to sit’, which
are, respectively, lētan and sēton in Northumbrian, Kentish and Mercian. However, together with the West-Saxon
variety, Northumbrian and Mercian also present some <æ> forms, as in dǣlan ‘to divide; to distribute’, hǣlan ‘to
heal’ or lǣdan ‘to lead’; while Kentish uses the <e> forms of these words: dēlan, hēlan and lēdan.
Another contrast holds between <ie> and <e, æ>. West-Saxon is the only variety of Old English which uses the
<ie> forms for verbs such as hliehhan ‘to laugh’, cierran ‘to turn’, hīeran, gelīefan ‘to believe’ and giefan ‘to give’;
while the <e> or <æ> forms of these words are preferred in Northumbrian, Kentish and Mercian: hlehhan/hlæhhan,
cerran, hēran, gelēfan and gefan.
West-Saxon also displays the diphthong <ea> where the other three varieties of Old English use <e> or <æ>.
This is the case with sceal ‘shall’ in West-Saxon, but scel or scæl in Northumbrian, Kentish and Mercian.
The contrast <eo>/<e> distinguishes West-Saxon and Kentish from Northumbrian and Mercian. The WestSaxon and Kentish varieties prefer the <eo> forms; hence verbs such as beorgan ‘to protect’ and flēogan ‘to fly’,
which are respectively bergan and flēgan in Northumbrian and Mercian.
<y> forms are used in all Old English dialects. However, in Kentish, the original <y> of these forms changes into
<e>. In this way, verbs like fyllan ‘to fill’ and ontynan ‘to open’ become fellan and ontenan in Late Kentish, although
the <y> forms continue to be used in the other dialects and Early Kentish texts.
The contrast <e>/<eo> distinguishes West-Saxon, which uses the <e> forms, from the rest of dialects, which
prefer the <eo> forms. An example of this is the verb beran ‘to bring’, which is beoran in Northumbrian, Kentish and
Mercian. Nevertheless, it should be noted that, together with West-Saxon, Northumbrian and Mercian also present
some <e> forms where Kentish still prefers the <eo> forms. Thus, the verb sprecan ‘to speak’ is written in this way
in all Old English dialects except Kentish, which favours spreocan.
Another contrast which characterizes the West-Saxon dialect of English is the /<io> contrast. forms are
preferred only in West-Saxon, whereas Northumbrian, Kentish and Mercian adopt <io> spellings. Therefore, the
West-Saxon forms sidu ‘habit’ and wita ‘adviser’ are siodu and wiota in the rest of the dialects.
The contrast <ea>/<a> also allows us to distinguish the West-Saxon variety, which presents <ea> forms, from
the other three, which use <a> spellings. Nevertheless, it should be noted that, in Kentish, the <a> becomes <ea>,
as in West-Saxon. Thus, healdan in West-Saxon and Late Kentish corresponds to haldan in Northumbrian, Mercian
and Early Kentish texts.
Another contrast holds between <ie> and <io, eo>. West-Saxon is characterized by using <ie> spellings where
the rest of dialects prefer the forms in <io> or <eo>. Hence, hierde ‘shepherd’ and gestrīenan ‘to procreate’ are
found in West-Saxon texts, but hiorde or heorde and gestrīona or gestrīonan are used in Northumbrian, Kentish
and Mercian.
Finally, another contrast obtains between <e> and <eo>. West-Saxon presents <e> forms where the rest of
the dialects opt for <oe> spellings. However, in Kentish the original <oe> spelling gives <e> forms. Thus, we find
dēman ‘to judge’ and sēcan ‘to seek’ in West-Saxon and late Kentish, but dōēman and sōēcan in Northumbrian,
Mercian and early Kentish. Similarly, ēþ ‘he does’ is used in West-Saxon and Late Kentish where Mercian and Early
Kentish use dōēþ. Nevertheless, this verbal form is doēs in Northumbrian.
Turning to intradialectal vocalic variation in West-Saxon, Early West-Saxon texts present <ie> forms, as in
hīeran ‘to hear’; hīerde, the preterite form of hīeran; begietst, the third person singular indicative of the verb
begietan ‘to obtain’; giefan ‘to give’; and wierþ, the third person singular indicative of the verb weorþan ‘to become’.
Nevertheless, this diphthong changes to <y> or . Therefore, in Late West-Saxon texts, we find hȳran or hīran,
hȳrde or hīrde, begystst or begitst, gyfan or gifan and wyrþ or wirþ instead of the spellings presented above.
RLyLA Vol. 09 (2014), 55-66 | 59
The contrast <y>/ can been identified between some nouns. In Early West-Saxon we find spellings like
cyning ‘king’, cynn ‘race’ and dryhten ‘lord’. However, this <y> evolved into . The corresponding forms in Late
West-Saxon are, respectively, cining, cinn and drihten.
In addition, Early West-Saxon presents <ea> forms for verbs such as reahte, the preterite of reccan ‘to narrate’;
seah, the preterite of sēon ‘to see’; geaf, the preterite of giefan ‘to give’; and sceal ‘shall’. This diphthong yields way
to <e>. Thus, in Late West-Saxon we find rehte, seh, gef and scel.
Another contrast can be identified between <io> and <eo>. Early West-Saxon displays <io> forms such as
cliopode, the preterite of the verb clipian ‘to call’; and liofast, the second person singular present indicative of the
verb libban ‘to live’. In contrast, in Late West-Saxon, these forms are cleopode and leofast respectively.
Among vocalic contrasts we must also include some represented by consonants, like the one holding between
<v[j]> and <v>. Thus, in Early West-Saxon we find forms like frignan ‘to ask’; ligeþ, the third person singular
present indicative of the verb licgan ‘to lie’; and sægde, the preterite of secgan ‘to say’, which become frīnan, līp
and sæde in Late West-Saxon.
Finally, other diachronic verbal contrasts are a consequence of the gradual regularization that Old English verbs
undergo throughout the process of simplification of inflections. In this way, some originally canonical forms of
strong verbs like sprecen, the present subjunctive plural of the verb sprecan ‘to speak’; and sungon, the preterite
indicative plural of the verb singan ‘to sing’, adopted a weakened form, namely sprecan and singan.
4. ALTERNATION Vs. VARIATION IN THE OLD ENGLISH DE TEMPORIBUS ANNI
This section shows the results evinced by the analysis of alternation and variation in De Temporibus Anni.
Beginning with alternations, this part of the analysis is based on the proposal by Kastovsky (1968), as presented
in section 3. The focus is on alternations, involving both short and long vowels, which hold between the strong
verb and the morphologically related weak verb. The alternations involving short vowels with instances in
De Temporibus Anni can be seen in (7):
(7) a.
<a> ~ <æ>
habban [habban inf; habbað pr3p; hæbbe subj pr3s; gehæfd pp; hæfð pr3s with neg; nabbað pr1p pr3p; næbbe subj pr3s; næfð pr3s] wk. 3 ‘have, possess, hold’ (habban ‘to have’)
b.
<ea> ~ <y> (=<ie>)
āfyllan [āfylled pp; āfyllede pp npm] wk. 1 ‘fill’ (feallan ‘to fall’)
c.
<eo> ~ <ie>
forbærnan (=forbiernan, Sweet) [forbærn sing imp; forbærne subj pr3s]
wk. 1 ‘burn, consume by fire’ (beornan ‘to burn’)
Likewise, (8) shows those alternations involving long vowels:
(8) a.
<ā> ~ <ǣ>
geneālǣcan [genealǣce pr3s] wk. 1 ‘approach’ (lācan ‘to move up and down’)
geþwǣrlǣcan [geðwǣrlācað pr3p] wk. 1 ‘agree, be in harmony’ (lācan ‘to move up and down’)
winterlǣcan [winterlǣcð pr3s] wk. 1 ‘grow wintry’ (lācan ‘to move up and down’)
b.
<ī> ~ <ǣ, preterite ā>
gelǣdan [gelǣd pp nsm] wk. 1 ‘lead, guide, conduct’ ((ge)līðan ‘go to’)
tōdrǣfan [todrǣfð pr3s] wk. 1 ‘scatter, disperse’ (drīfan ‘drive’)
c.
<ō> ~ <ē>
oferflēdan [oferflēt pr3s] wk. 1 ‘overflow, flood’ (flōwan ‘to flow’)
d.
<ēo> ~ <īe>
āflīegan [āflīgð pr3s] wk. 1 ‘drive away, put to flight’ (flēogan ‘to fly’)
e.
<ēo> ~ <ȳ> (=<īe>)
ālȳsan [ālȳsed pp] wk. 1 ‘set free, release’ (lēosan)
f.
<ū> ~ <ī> (=<ȳ>)
gebīgan [gebīgedum pp dpn] wk. 1 ‘bend’ (būgan ‘bow’)
RLyLA Vol. 09 (2014), 55-66 | 60
As shown in figures 1 and 2, Kastovsky (1968) does not distinguish the alternation A4AR, involving <e> and
, in his study of strong verb-noun alternations. However, in the analysis of De Temporibus Anni, this alternation
holds between the strong and the weak verb, as is shown in (9):
(9) <e> ~ 
āwendan [awend pp; awent pr3s] wk. 1 ‘turn, turn aside, change’ (windan ‘to wind’)
(ge)settan [geset pp nsm; gesett pp nsm nsf; gesette pt3s; setton pt3p; gesetton pt3p] wk.1 ‘to establish,
compile, allocate’ (sittan ‘to sit’)
wendan [wendað pr3p; wende subj pr3s; went pr3s] wk. 1 ‘turn, go, proceed’ (windan ‘to wind’)
gewendan [gewent pr3s] wk. 1 ‘return’ (windan ‘to wind’)
Evidence of the type provided in (9) reinforces the systematic and predictable character of alternations, since,
given a general framework mainly based on the evolution of i-mutation, it is possible to fill in its blanks in such a
way that the main principles on which the classification is based are kept. In other words, by selecting the class
of the weak verb, it is possible to find the reversal of the alternation in the direction of /i/, which obtains when the
strong verb is the base of nominal derivation.
Turning to variation, the results thrown by the analysis have been classified by contrast. The classification of
interdialectal contrasts is provided in (10). It should be noted that interdialectal contrasts are presented by taking the
West-Saxon spellings as reference. Thus, in the description of the contrasts in (10), the form corresponding to the
West-Saxon dialect appears in the first place. Verbal forms have been divided into predictable and unpredictable
depending on the spelling they present in De Temporibus Anni. The first group of verbs of each contrast corresponds
to those verbs with a form compatible with the West-Saxon dialect. Conversely, unpredictable contrasts refer to
those verbs which do not present the West-Saxon form. The spellings appearing in De Temporibus Anni are shown
at the left throughout the classification.
(10)a. <æ>/<e>
Predictable
bedæled [bedælan] ~ bedeled
betæhte [betæcan] ~ betec
gefæstnod [gefæstnian] ~ gefestnie
forbærn, forbærne [forbærnan] ~ forbernan
Unpredictable
acenð, acennedum [acennan] ~ acænnan
asendan [asendan] ~ asændan
aðenede [aðennan] ~ aþænede
awend, awent [awendan] ~ awændan
derað [derian] ~ dærigen
geendað, geendod, geendode [geendian] ~ geændian
fremað [fremian] ~ fræmeð
b. <ie>/<e, æ>
Predictable
afligð [afliegan] ~ aflegedo
Unpredictable
aberst [aberstan] ~ abiersð
berð [beran] ~ viere
c. <ea>/<e, æ>
Predictable
afeallað [afeallan] ~ afellan
aheawene [aheawan] ~ ahewenne, ahæwenum
feallað, fealð, fylð [feallan] ~ fellan
Unpredictable
berð [beran] ~ beara
betæhte [betæcan] ~ beteahte
gefæstnod [gefæstnian] ~ gefeastnadon
forbærn, forbærne [forbærnan] ~ forbearnde
RLyLA Vol. 09 (2014), 55-66 | 61
d.
<eo>/<e> (West-Saxon <eo> corresponds to Germanic [eu])
Predictable
aðeostrian, aðeostrað [aðeostrian] ~ aðestred
beboden [bebeodan] ~ bebed
bescyt [besceotan] ~ bescet
fleon, fleoð [fleon] ~ flegan
e. <y>/<e>
Predictable
adylegode [adylegian] ~ adelegað
afylled, afyllede [afyllan] ~ afelle
alysed [alysan] ~ alesan
astyrað, astyred [astyrian] ~ astereð
gebyrige [gebyrian] ~ gebereð
cyrð, gecyrð [(ge)cyrran] ~ gecerran
gefyllað [gefyllan] ~ gefellan
Unpredictable
aberan, aberð [aberan] ~ abyrð
aberst [aberstan] ~ abyrst
abrece [abrecan] ~ abrycan
acenð, acennedum [acennan] ~ acynð
berð [beran] ~ byran
derað [derian] ~ dyrige
f. <e>/<eo> (West-Saxon <e> corresponds to Germanic [e])
Predictable
aberan, aberð [aberan] ~ abeoren
atent [atendan] ~ ateodon
berð [beran] ~ beoran
derað [derian] ~ deoriende
Unpredictable
awyrpð [aweorpan] ~ awerpan
g. /<io>
Predictable
arisan, arison, arist [arisan] ~ arioson
geedniwod [edniwian] ~ edniowað
h. <ea>/<a>
Predictable
afeallað [afeallan] ~ afalle
behealdan [behealdan] ~ behaldan
feallað, fealð, fylð [feallan] ~ fallen
Unpredictable
beheton [behatan] ~ beheatenre
gegaderað, gegaderode, gadrian [(ge)gad(e)rian] ~ gegeadriga
i. <ie>/<io, eo>
Predictable
aðeostrian, aðeostrað [aðeostrian] ~ aðiestrige
Unpredictable
æteowað, æteowiað [æteowian] ~ ætiewan
beboden [bebeodan] ~ bebiet
fleon, fleoð [fleon] ~ fliehð
j. <e>/<oe>
Predictable
awend, awent [awendan] ~ awoendað
Concerning intradialectal contrasts, it should be taken into account that its analysis is circumscribed to the
West-Saxon dialect. In this way, in (11), intradialectal contrasts are identified between Early and Late West-Saxon
forms. Early West-Saxon graphemes are presented in the first place in the description of each contrast. Under the
category of predictable, (11) displays those verbs with Late West-Saxon forms in De Temporibus Anni, while Early
West-Saxon spellings have been grouped under the category of unpredictable.
RLyLA Vol. 09 (2014), 55-66 | 62
(11)a.
<ie>/<y, i>
Predictable
afylled, afyllede [afyllan] ~ afielde
alysed [alysan] ~ aliesan
awrat, awritenne, awriton [awritan] ~ awrieten
gebicniað [bicnian] ~ biecne
gebigedum [gebigan] ~ gebiegeð
cyrð, gecyrð [(ge)cyrran] ~ gecierran
gefyllað [gefyllan] ~ gefielde
Unpredictable
afligð [afliegan] ~ aflygan, afligan
b. <y>/
Predictable
abæd, abæde [abiddan] ~ abyddan
afindan [afindan] ~ afynden
agifð [agifan] ~ agyfan
arisan, arison, arist [arisan] ~ arysan
astah, astihð [astigan] ~ astygen
awrat, awritenne, awriton [awritan] ~ awrytan
belicð [belicgan] ~ bilyð
belimpað, belimpð [belimpan] ~ belympð
gebicniað [bicnian] ~ bycneþ
bæd [biddan] ~ byddan
gebæd [gebiddan] ~ gebyddan
gebigedum [gebigan] ~ gebygan
gediht [dihtan] ~ dyht
geedniwod [edniwian] ~ ednywod
forgifan, forgeaf [forgifan] ~ forgyfan
glit [glidan] ~ glyt
Unpredictable
adylegode [adylegian] ~ adilegian
afylled, afyllede [afyllan] ~ afillað
alysed [alysan] ~ alisan
astyrað, astyred [astyrian] ~ astirian
gebyrige [gebyrian] ~ gebireþ
cyrð, gecyrð [(ge)cyrran] ~ gecirran
fyligð [fyligan] ~ fulfiligan
gefyllað [gefyllan] ~ gefillan
c. <ea>/<e>
Predictable
berð [beran] ~ beara
Unpredictable
afeallað [afeallan] ~ afellan
aheawene [aheawan] ~ ahewenne
eardað [eardian] ~ erddian
feallað, fealð, fylð [feallan] ~ fellan
d. <io>/<eo>
Predictable
atihð [ateon] ~ ation
aðeostrian, aðeostrað [aðeostrian] ~ aþiostraþ
awyrpð [aweorpan] ~ æwiorpen
æteowað, æteowiað [æteowian] ~ atiowan
beboden [bebeodan] ~ bebiode
fleon, fleoð [fleon] ~ flion
Table 1 and 2 provide the figure of instances of interdialectal and intradialectal contrasts respectively. The
results of analysis are presented by contrast. The classification into predictable and unpredictable follows the same
criteria as in the analysis above. In table 1, which is devoted to interdialectal contrasts, the column of predictable
instances displays the figure of verbs with West-Saxon spellings in De Temporibus Anni, whereas the column of
unpredictable instances provides the number of verbs which present a Kentish, Mercian or Northumbrian form.
RLyLA Vol. 09 (2014), 55-66 | 63
Similarly, the column of predictable instances in table 2 shows the figure of verbs that present Late West-Saxon
forms; whereas the column of unpredictable instances gives the number of verbs showing spellings identified as
Early West-Saxon.
Table 1. Interdialectal contrasts.
Type of contrasts
Interdialectal
<æ>/<e>
Number of instances
Predictable
Unpredictable
4
7
<ie>/<e, æ>
1
2
<ea>/<e, æ>
3
4
<eo>/<e>
4
0
<y>/<e>
7
6
<e>/<eo>
4
1
/<io>
2
0
<ea>/<a>
3
2
<ie>/<io, eo>
1
3
<e>/<oe>
1
0
Total
30
25
Table 2. Intradialectal contrasts.
Type of contrasts
Intradialectal
Number of instances
Predictable
Unpredictable
<ie>/<y, i>
7
1
<y>/
16
8
<ea>/<e>
1
4
<io>/<eo>
6
0
Total
30
13
As can be seen in tables 1 and 2, the instances of interdialectal contrasts outnumber those of intradialectal
ones. To be more precise, a total of 55 interdialectal and 43 intradialectal contrasts have been identified in De
Temporibus Anni. Moreover, the figures of instances of interdialectal and intradialectal contrast vary considerably.
With regard to interdialectal contrasts, the contrasts which present the highest number of instances are <y>/<e>,
<æ>/<e> and <ea>/<e, æ>, with 13, 11 and 7 instances respectively. Conversely, the least frequent interdialectal
contrasts are <e>/<oe>, /<io> and <ie>/<e, æ>, which present 1, 2 and 3 instances respectively. Similarly, there
is a high degree of variation regarding the number of instances of the different intradialectal contrasts, <y>/
being the one which presents the highest number of instances – 24 – and <ea>/<e> the least frequent with 5
instances. Furthermore, out of the 50 verbs which present at least one contrast, 28 have instances of two or more
different contrasts. In interdialectal analysis, it turns out that the contrasts <æ>/<e>, <ie>/<e, æ>, <ea>/<e, æ>
and <ie>/<io, eo> show more unpredictable instances than predictable ones. In intradialectal analysis, the contrast
<ea>/<e> has more unpredictable than predictable analysis. Overall, there are more instances of predictability than
of unpredictability, although the degree of unpredictability is very high, considering that the analysis is restricted
to one text. The data, therefore, clearly indicate that variation, unlike alternation, is a relatively unsystematic and
unpredictable phenomenon.
5. CONCLUSION
This article has analyzed morphophonological alternations as well as diatopic and diachronic variation as shown
in the Old English version of De Temporibus Anni. The analysis has stressed the importance of the diphthong <ie>
not only as a criterion for identifying early West-Saxon but also as a point of contact with the phenomenon of
i-mutation, which constitutes the diachronic motivation of the direct and reverse vocalic alternations proposed
by Kastovsky. Regarding the aim of drawing a distinction between the concepts of alternation and variation, the
RLyLA Vol. 09 (2014), 55-66 | 64
analysis that has been carried out indicates that the phenomenon of alternation is relatively predictable, relatively
systematic and has a tendency to be generalized, in contradistinction to variation, which is relatively unpredictable
and unsystematic and tends to be local. On the descriptive side, the alternation A4aR <e> ~ holds between
the strong verb and the weak verb. The alternation A11R <ū> ~ <ȳ> has not been found in the text under analysis,
thus representing a pending task for future research.
REFERENCES
Bammesberger, A. (1965). Deverbative jan-Verba des Altenglischen, vergleichend mit den übrigen altgermanischen
Sprachen dargestellt. München: Ludwig-Maximilians Universität.
Blake, M. (2009). Ælfric’s de Temporibus Anni. Cambridge: D. S. Brewer.
Hinderling, R. (1967). Studien zu den starken Verbalabstrakten des Germanischen. Berlin: Walter de Gruyter.
doi:10.1515/9783110822663
Hogg, R. M. (1992). “Phonology and Morphology”. The Cambridge History of the English Language. Volume I: The
Beginnings to 1066. R. M. Hogg (ed.). Cambridge: Cambridge University Press. pp. 67-167.
Hogg, R. M. and R. D. Fulk. (2011). A Grammar of Old English. Volume 2: Morphology. Oxford: Blackwell.
Kastovsky, D. (1968). Old English Deverbal Substantives Derived by means of a Zero Morpheme. PhD Dissertation.
Tübingen: Eberhard-Karls-Universität.
Kastovsky, D. (1986). "Deverbal nouns in Old and Modern English: from stem-formation to word-formation". In J.
Fisiak (ed.), Historical Semantics-Historical Word Formation. Berlin: Mouton de Gruyter. pp. 221-261.
Kastovsky, D. (1989). "Morphophonemic alternations and the history of English: Examples from Old English".
In Manfred Markus (ed.), Historical English. On the occasion of Karl Brunner’s 100th birthday. Innsbruck:
Universität. pp. 112-123.
Kastovsky, D. (1990). "The typological status of Old English Word Formation". In S. M. Adamson, V. A. Law, N.
Vincent and S. Wright (eds.), Papers from the 5th International Conference on English Historical Linguistics.
Amsterdam: John Benjamins. pp. 205-224. doi:10.1075/cilt.65.14kas
Kastovsky, D. (1992). "Semantics and vocabulary". In R. M. Hogg (ed.), The Cambridge History of the English
Language I: The Beginnings to 1066. Cambridge: Cambridge University Press. pp. 290-408. doi:10.1017/
CHOL9780521264747.006
Kastovsky, D. (2006). "Typological Changes in Derivational Morphology". In A. van Kemenade and B. Los (eds.), The
Handbook of the History of English. Oxford: Blackwell publishing. pp. 151-177. doi:10.1002/9780470757048.
ch7
Lass, R. and J. Anderson. (1975). Old English Phonology. Cambridge: Cambridge University Press
Martín Arista, J. (2008). "Unification and separation in a functional theory of morphology". In R. D. Van Valin, Jr.
(ed.), Investigations of the Syntax-Semantics-Pragmatics Interface. Amsterdam: John Benjamins. pp. 119145. doi:10.1075/slcs.105.12mar
Martín Arista, J. (2009). "A Typology of Morphological Constructions". In C. Butler and J. Martín Arista (eds.),
Martín Arista, J. (2010a). "OE strong verbs derived from strong verbs". SKASE Journal of Theoretical Linguistics,
7: 36-56.
Martín Arista, J. (2010b). "Lexical negation in Old English". NOWELE (North-Western European Language
Evolution), 60/61: 89-108.
Martín Arista, J. (2011a). "Morphological relatedness and zero alternation in Old English". In P. Guerrero Medina
(ed.) Morphosyntactic Alternations in English. Functional and Cognitive Perspectives. Sheffield: Equinox.
pp. 339-362.
Martín Arista, J. (2011b). "Adjective formation and lexical layers in Old English". English Studies, 92/3: 323-334.
doi:10.1080/0013838X.2011.564776
Martín Arista, J. (2011c). "Projections and constructions in functional morphology: the case of HRĒOW". Language
and Linguistics 12/2: 393-425.
Martín Arista, J. (2012a). "The Old English Prefix ge-: A Panchronic Reappraisal". Australian Journal of Linguistics,
32/4: 411-433. doi:10.1080/07268602.2012.744264
Martín Arista, J. (2012b). "Morfología flexiva en RRG". In R. Mairal, L. Guerrero and C. González (eds.), El
Martín Arista, J. (2013). "Recursivity, derivational depth and the search for Old English lexical primes". Studia
Neophilologica, 85/1: 1-21.
Martín Arista, J. "Noun layers in Old English. Inheritance and mismatches in lexical derivation". Forthcoming-a.
RLyLA Vol. 09 (2014), 55-66 | 65
Martín Arista, J. "El paradigma derivativo del inglés antiguo". Forthcoming-b.
Palmgren, C. (1904). English Gradation-Nouns in Their Relation to Strong Verbs. Uppsala: University of Uppsala.
Ray, B. K. (1996). Old English Morphology and Indo-European. Delhi: Sharada Publishing House.
Ringe, D. (2006). From Proto-Indo-European to Proto-Germanic. A Linguistic History of English. Volume I. Oxford:
Oxford University Press. doi:10.1093/acprof:oso/9780199284139.001.0001
Rix, H., M. Kümmel, T. Zenhder, R. Lipp and B. Schirmer. (2001). Lexicon der indogermanischen Verben. Wiesbaden:
Ludwig Reichert Verlag.
Schön, E. (1905). Die Bildung des Adjectivs im Altenglischen. Kiel: Verlag von Robert Cordes.
Schuldt, C. (1905). Die Bildung der schwachen Verba im Altenglischen. Kiel: Verlag von Robert Cordes.
Szemerényi, O. (1996). Introduction to Indo-European Linguistics. Oxford: Oxford University Press.
RLyLA Vol. 09 (2014), 55-66 | 66
Vol. 9 año 2014, 67-78
EISSN 1886-6298
CAN LEXICAL ERRORS INFORM ABOUT WORD CLASS ACQUISITION IN THE FOREIGN
LANGUAGE?: EVIDENCE FROM GREEK LEARNERS OF SPANISH AS A FOREIGN LANGUAGE
Kiriakí Palapanidi
Universidad Abierta de Grecia
María Pilar Agustín Llach
Universidad de la Rioja
Abstract: This research study examines the order of acquisition of four different classes of words: verbs, nouns, adjectives,
and adverbs in three different levels of proficiency in the FL. We apply the procedure of Error Analysis to a corpus of written
compositions and analyze the interlingual and intralingual lexical errors (formal and semantic) of Greek students studying Spanish
as a Foreign Language (SFL). Our results have shown i) that there is a relationship between the type of word and the production of
lexical errors and ii) that the evolution of lexical errors presents different trends for different word classes. Additionally, our results
lead us to the conclusion that word class is a stronger predictor of lexical error type than proficiency level.
Keywords: error analysis, lexical errors, word class acquisition in FL, vocabulary acquisition in FL, acquisition of Spanish as a FL.
1. INTRODUCTION
Grammar distinguishes two types of words: grammatical and lexical words. Grammatical words are the ones
which are used to express grammatical relations, traditionally articles, prepositions, pronouns, and conjunctions.
On their part, lexical words are defined as the ones that express meaning or contribute to providing speech with
meaning. These are: nouns, verbs, adjectives and adverbs. There may be other classifications of lexical items,
but this is the taxonomy most frequently acknowledged in the literature. In the present paper, we will concentrate
exclusively on lexical words. Traditionally, there is the firm belief that all types of words are not equally easy to
acquire. This belief has its origin in the observation of native language acquisition (Caselli et al., 1995; Bornstein
et al., 2004; Gentner and Baroditsky, 2009), where children seem to acquire nouns first, then verbs, and finally
adjectives and adverbs, respectively.
Frequency of input, lack of cognitive complexity, which is linked to the quick and easy identification of nouns
in real life, those with concrete referents, especially; and conceptual development are the factors alluded to when
accounting for order of acquisition of word classes. Verbs are more difficult to acquire than nouns, because they
depend on some language-specific knowledge related to syntax and semantic components, such as motion,
manner, and other aspects encoded in verbs; and require thus a higher cognitive load (Naigles and Hoff-Ginsberg
1998, Myles 2004, VanPatten 2004). Finally, adjectives and adverbs appear to qualify the nouns, the actions or the
processes expressed. They are not always necessary to get the message across, but contribute to making it clear
and to eliminating possible ambiguities or misunderstandings.
Although it is difficult to quantify, few studies have addressed the issue of which types of words are acquired
first and last in the process of foreign or second language acquisition (Schmitt 1998, San Mateo Valdehíta
2003/2004 for Spanish FL). Research notes that nouns and fixed expressions come first in L2 acquisition. In
particular, concrete nouns and those easily visualized are learned before abstract ones (Sökmen 1997). Similarly,
cognates are also easy and appear early in L2 productions (Jarvis 2000). In studies about lexical availability in
Spanish FL, the most frequently available words are nouns, followed by verbs and adjectives in varying positions
depending on the semantic field under scrutiny (Šifrar Kalan 2009). Housen et al. (2008) found that adolescents
produce more verbs than nouns in French L2, whereas David (2008) found nouns to be the first and most frequent
| 67
Kiriakí Palapanidi and María Pilar Agustín Llach
Can lexical errors inform about word class acquisition in the foreign language?:
Evidence from Greek learners of Spanish as a foreign language
category among younger learners in their oral French L2 discourse, with verbs increasing faster than nouns as
learners go up grade.
The early appearance of, more or less, fixed expressions is accounted for by alluding to the assumption that,
because they make up a single unit of meaning, they are learned together as if they were a single lexical item.
These fixed expressions have been proved highly effective in communication (Wray and Fitzpatrick 2010). This
is very much in the line of the construction theory within the field of cognitive linguistics. This field of study
claims that language is made up of chunks or constructions which constitute a form-meaning unit. Accordingly,
L2 learning consists basically in learning different constructions (Robinson and Ellis 2008). The pre-eminence of
semi-fixed patterns in early language makes a strong claim for this theory. Chunks are a frequent phenomenon in
the productions of young and beginner learners, who are unable to analyse language and linguistic structures and
therefore, dissect them into smaller components.
Additionally, research has evidenced that in mother tongue acquisition nouns are the first words to be learned
when they refer to realities in the child’s world. For instance, according to Gentner and Baroditsky (2009), in the
productive vocabulary of the Navajo children, nouns such as those for animals (cat, dog), food (banana, orange,
water), or people (grandmother, father, brother, teacher) predominate in their productions. Verbs come immediately
after, since they represent actions, or processes needed to link nouns, and therefore are basic for the development
of syntax, i.e. to construe meaningful and logical sentences. Adjectives and adverbs appear last, since they are not
indispensable to transmit content, although on most occasions they modify or complement the content meaning
of nouns or verbs. In general, research with different target languages of Indo-European origin coincide with the
same order of acquisition observed when these languages are acquired as a mother tongue (Gentner 1982, Tardif,
Shatz, Naigles 1997).
Nevertheless, some years ago, researchers working with L1 development questioned the nouns first hypothesis
on the basis of evidence from other non-Indo-European languages (Tardif, Shatz, Naigles 1997, Naigles and HoffGinsberg 1998). The salience of verbs in the input of some languages lies at the origin of this reservation and
questioning. Indeed, according to the usage based model (e.g. Diessel 2004); the frequency of occurrence of a
linguistic element makes its activation easier and quicker. According to this, the elements most frequent or most
salient in the input would be acquired most easily. However, research in this respect is scarce and no definite
conclusions have been reached. In the present study, we are interested in going deeper into the issue of the
order of word class acquisition by examining the lexical errors in each of the categories. This represents a novel
perspective in this type of studies, because former research on word class acquisition has not used error analysis
as a methodology. Furthermore, studies on lexical errors do not tend to classify the errors identified according to
word classes (but cf. Lennon 1991).
2. LEXICAL ERRORS AND LANGUAGE ACQUISITION
According to Myles (2004), the linguistic forms (correct and incorrect) produced by foreign language (FL)
learners are a window into their mental representations, and hence they serve as indicators of the development of
the language acquisition process. According to Agustín Llach and Barreras Gómez (2007), lexical errors represent
a hint into vocabulary acquisition in an FL given the fact that they may demonstrate a lack of lexical knowledge,
possible communication strategies used by the learners, the organization of the mental lexicon, and/or the
development of lexical competence. In this sense, lexical errors can be very useful in helping to delve into the
issue of the order of word class acquisition.
Notwithstanding the above claims, Myles (2004) believes that the relationship between the produced linguistic
forms, whether they are correct or erroneous, and the mental representations of the learners is not simple and
direct. This relationship is more complicated and multidimensional. For instance, lack of lexical errors of a particular
word class, or linguistic item can be the result of mastery but also of inhibition in use, a phenomenon generally
known as avoidance (Schachter 1974).
Although it is difficult to identify the underlying cause of a wrong linguistic form, lexical errors can be very
valuable in getting a deeper, clearer, and more accurate picture of the processes of lexical acquisition. Moreover,
the positive view of the (lexical) errors has generally been agreed on since Corder (1967, 1973), who claims that
(lexical) errors are considered a sign of language acquisition and development rather than lack thereof.
RLyLA Vol. 09 (2014), 67-78 | 68
3. LEXICAL ERRORS IN THE WRITTEN COMPOSITIONS OF GREEK SFL LEARNERS
With these considerations in mind, this study intends to explore the order of the acquisition of words of four
different classes: verbs, nouns, adjectives, and adverbs by detecting the class of word that causes the most
difficulties at the different levels of proficiency in the FL. In order to accomplish this objective, we apply the
procedure of Error Analysis to a corpus of written compositions and analyze the interlingual and intralingual lexical
errors (formal and semantic) of Greek students studying Spanish as a Foreign Language (SFL).
Based on previous research, we are led to believe that lexical errors in nouns and verbs would decrease with
increasing proficiency, but errors in adjectives and especially in adverbs would increase as learners proficiency
progresses from beginner to intermediate or advanced. Our starting point is the assumption that learners master
nouns and verbs before they get to master adjectives and adverbs whose presence augments at higher proficiency
levels. We believe that examination of lexical errors will help us throw further light in this issue. We were also
interested in examining the types of lexical errors most frequent for each word class and thus in identifying any
systematicity in their production that might help unveil the process of acquisition of the specific word classes.
In order to explore the nature of the relationship between word class, lexical error production and proficiency
level, in the present study we set out to investigate the following research questions:
1. Is there evidence of a relationship between word class and the production of lexical errors?
2. Does this relationship change with FL proficiency?
3. What is the most frequent lexical error type in nouns, verbs, adjectives, and adverbs?
4. Do types of lexical errors in word class change as proficiency increases?
4. METHOD
4.1. Informants
Our sample is composed of 119 Greek SFL learners who have been classified into three different linguistic
levels of proficiency: B1, B2, and C1 as defined by the European Reference Framework. Students were classified
into the proficiency levels according to the result of a placement test, the Intermediate DELE1 Exam of November
2008. Thus, those who obtained between 30 and 40 correct answers were ascribed to the B1 level, between 40
and 50 to the B2 and above 50 to the C1 level.
According to this, we assigned 32 students in level B1, 47 students in level B2, and 40 students in level C1. All
the participants were adults, of Greek mother tongue, and they were learning SFL in an institutionalized context2.
Table 1. Summary of informants ascribed to proficiency level.
No. of students
B1
32
B2
47
C1
40
4.2. Instruments of data collection
The topic of the composition was the narration of a film that they had seen recently (Figure 1). A general and
simple topic was chosen which allowed students from all three proficiency levels to complete the writing task
successfully. No maximum or minimum word or length constraints were imposed. Participants were not allowed
to use any support material such as dictionaries or grammar reference books. They were allotted 30 minutes to
complete the writing task.
1) Probablemente eres aficionado al cine. ¿Recuerdas la última película que has visto? Cuéntala y habla de:
dónde, cuándo y cómo se inicia la historia; qué sucede posteriormente; cuál es el momento más importante;
cómo termina.
Figure 1. Composition Topic.
1
2
DELE is the Spanish acronym which stands for Spanish as a Foreign Language Certificate. It is the official examination of the Instituto Cervantes.
23 of our participants are Greek students of the Department of Spanish Literature of the National University of Athens and 96 of our participants are Greek SFL students
of the Center of Foreign Languages of the National University of Athens.
RLyLA Vol. 09 (2014), 67-78 | 69
4.3. Procedures and analysis
Students wrote the essay in regular classes with no extra help from dictionaries or grammar reference books,
as stated above. After writing the composition, they were asked to take the placement exam. The subsequent
phase in the study consisted in the identification and the classification of the lexical errors found. Taking into
account the objective of this study and based on previous studies on lexical errors (Zimmerman, 1987; Fernández,
1997; Jodar, 2006; Palapanidi, 2012), their lexical errors were classified into formal and semantic. In addition to
this, lexical errors were also classified into interlingual, or intralingual based on the criterion of the psychological
origin of the lexical error, which has been used by previous studies of a similar type (Zimmerman, 1987; Vázquez,
1991; Palapanidi, 2012).
The category of lexical interlingual formal errors includes the following errors subtypes (see also Figure 2):
1. gender (la comportamiento for el comportamiento)
2. number (los gentes for la gente)
3. adaptations from other foreign languages (el senario for el guión)
4. change of code, that is the learner uses a word from the L1 or other known languages (su atelier for taller)
Whereas, the category of lexical interlingual semantic errors is further subdivided into:
5. lexemes with common semes but not interchangeable in the context (hacer paseos for dar paseos)
6. literal translation (la primera actriz for la protagonista)
7. false friends (el protagonista realició for se dio cuenta)
8. use of inappropriate register (estación de téle for television)
9. use of circumlocution (tan mucho ruido for mucho ruido)
The category of lexical intralingual formal errors includes the following types of errors:
10. gender (muchas problemas for muchos problemas)
11. number (da comidas y información for comida)
12. use of formally similar target language words (su viaje madre for su madre vieja)
13. creation of non-existent words (una vida más sanida for sana)
On its part, the category of lexical intralingual semantic errors includes:
14. use of circumlocution (segunda guerra del Mundo for segunda guerra mundial)
15. changes in derivatives of the same root (del muerto de su madre for de la muerte de su madre)
16. lexemes with common semes but not interchangeable in context (la gente no puede mover for no puede
transportarse)
17. errors in the use of ser and estar (no está un problema for no es un problema)
18. use of inappropriate register (ir de compras me da un humor de perros for no me gusta ir de compras)
19. errors in collocations (en concluso for en conclusion)
20. pleonasm, that is the learners use two words with the same meaning to reinforce the utterance (lo más
mejor for mejor).
Finally, all lexical errors in the present study have been classified according to the word class they belong to,
and thus:
•
verbs (la gente comen for come)
•
nouns (un ciudad for una ciudad)
•
adjectives (pequeño ciudad for pequeña)
•
adverbs (enferma en su mente for enferma mentalmente)
RLyLA Vol. 09 (2014), 67-78 | 70
Lexical interlingual errors
Formal
1. Gender.
2. Number.
3. Adaptations of other foreign languages.
4. Change of code.
Semantic
1. Lexemes with common semes but not
interchangeable in the context.
2. Literal translation.
3. False friends.
4. Use of inappropriate register.
5. Use of circumlocution.
Lexical intralingual errors
Formal
1. Gender.
2. Number.
3. Use of formally similar target language words.
4. Creation of non-existent words.
Semantic
1. Use of circumlocution.
2. Changes in derivatives of the same root.
3. Lexemes with common semes but not
interchangeable in the context.
4. Errors in the use of ser and estar.
5. Use of inappropriate register.
6. Errors in collocations.
Figure 2. Typology of lexical errors.
5. RESULTS
In the first place, we were interested in exploring the relationship between lexical error production and word
class. Of the total of 511 lexical errors that have been identified in our corpus, 137 errors were committed in
verbs, which make up an average per participant of 1.15 errors, 272 in nouns with an average of 2.28 errors per
participant. We identified 89 errors in adjectives with an average of 0.75 errors per participant; and finally, 13 lexical
errors in adverbs with an average of 0.11 errors per participant.
In general, therefore, we see nouns as the word class where most lexical errors are produced, followed by
verbs, adjectives and adverbs in this order. Lexical errors in adverbs are very few.
Table 2 offers the figures for raw lexical error production, mean productions, and percentage of errors in each
word class over total across the three proficiency levels within the study. The table shows thus the evolution in the
production of lexical errors for each word class as proficiency increases.
Table 2. Evolution of lexical error production in the different word classes across level.
Word Class
Verb
Noun
Adjective
ProficLevel
No
Av
% over
total of
errors
B1
47
1.47
33.1
73
2.28
51.41
19
0.59
13.38
3
0.09
2.11
B2
63
1.34
26.92
127
2.7
54.27
42
0.89
17.95
2
0.04
0.85
C1
27
0.68
20
72
1.8
53.33
28
0.7
20.74
8
0.2
5.93
No.
Av.
% over
total of
errors
No.
Adverb
Av.
% over
total of
errors
No.
Av.
% over
total of
errors
If we examine general lexical error production across levels, as our second research question posed, we
observe that at the B1 level, lexical errors in nouns are 73, 2.28 on average. This constitutes 51.41% of total lexical
errors. For verbs the figures are 47 instances of lexical error, with an average of 1.47 and 33.1% of total lexical
errors. We have identified 19 lexical errors in adjectives with an average of 0.59 errors and 13.38% of total lexical
errors. Finally, lexical errors in adverbs are the least frequent with a total of 3 errors, an average of 0.09, and 2.11%
over total lexical errors.
RLyLA Vol. 09 (2014), 67-78 | 71
Regarding the B2 level, the results of our analysis show that there were 127 lexical errors in nouns, with an
average of 2.7 errors, which constitutes the 54.27% of total of lexical errors. A total of 63 errors were committed in
verbs with an average of 1.34 errors, 26.92% of the total lexical errors. There were 42 lexical errors in adjectives,
with an average of 0.89 errors, 17.95% of total lexical errors. Finally, similar to the B1 level results, lexical errors
in adverbs present low numbers, with a total of 2 errors, with an average of 0.04 errors and 0.85% of total lexical
errors.
At the C1 level, our analysis reveals that there were 72 lexical errors in nouns with an average of 1.8 errors
per participant, 53.33% of total lexical errors. For the verbs, 27 lexical errors were found, with an average of 0.68
errors, which constitutes 20% of total lexical errors. As far as the adjectives, there were 28 errors with an average
of 0.7 errors and 20.74% of total lexical errors. There were 8 lexical errors in adverbs with an average of 0.2 per
participant and they constitute 5.93% of total lexical errors.
Thus, we observe that as learner proficiency increases, lexical errors in verbs tend to decrease, whereas lexical
errors in nouns present a decreasing tendency, with a slight increase at the B2 level and a final decline at the C1
level. It is worth noticing that they are always the most numerous, accounting for around half of all lexical errors.
Lexical errors in adjectives tend to increase and finally, lexical errors in adverbs also tend to increase from the B1
to the C1 level.
Nevertheless, taking into account that the frequency of use of each word class is different, we examine our
results in relation with the number of produced verbs, nouns, adjectives and adverbs.
Table 3 presents the absolute numbers of produced verbs, nouns, adjectives and adverbs and their mean
values. In addition to these figures, Table 3 also offers the percentage of errors in each word class over total of
produced verbs, nouns, adjectives and adverbs across the three proficiency levels.
Table 3. Total production of every word class and percentage of errors over
the total production of every word class across proficiency level.
Word Class
Verb
Noun
Adjective
Pr.L.
No
Av
% over
total of
verbs
B1
712
22.25
6.6
888
27.75 8.22
294
9.19
6.46
118
3.69
2.54
B2
1174
24.98
5.37
1767
37.6
7.18
585
12.45
7.18
174
3.7
1.15
C1
1175
29.38
2.3
1632
40.8
4.41
580
14.5
4.83
179
4.48
4.47
No
Av
% over
total of
nouns No
Adverb
Av
% over
total of
adj.
No
Av
% over
total of
adv.
The results of our analysis indicate that the most used word class is nouns in all proficiency levels, followed
by verbs, adjectives and adverbs, in this order. This last category has a very weak presence. Furthermore, our
analysis based on the total production of every word class verifies the strong presence of lexical errors in nouns.
We found high percentages of lexical errors in nouns when the total production of nouns used by the participants
in all proficiency levels was considered.
Regarding the evolution of lexical errors in the four word classes, the results of our analysis show that at the
B1 level the percentage of lexical errors in nouns over the total of nouns produced is 8.22%, for verbs we find
a percentage of 6.6%, for adjectives 6.46% and for adverbs 3.39%. At the B2 level, our analysis indicates a
percentage of 7.18% of nouns affected by a lexical error, 7.18% of the adjectives, 5.37% of the verbs and 1.15%
of the adverbs. Finally at the C1 level, we observe a percentage of 4.41% nouns affected, 2.3% verbs, 4.83%
adjectives and 4.47% adverbs.
Thus, we can see that these results verify the decreasing tendency of lexical errors in nouns and verbs and the
increase of the lexical errors in adjectives and adverbs.
In order to answer our third and fourth research questions, we analyzed lexical error categories to find out
the most frequent lexical error type for each of the different word classes generally considered and across the
proficiency levels.
RLyLA Vol. 09 (2014), 67-78 | 72
Table 4 displays the mean values and absolute numbers of lexical error types as concerns their distribution
across the four word classes.
Table 4. Distribution of the lexical errors into the different word classes.
Word Class
Verb
Noun
Adjective
Adverb
Type of lexical error
No.
Av.
S.D.
No.
Av.
S.D.
No.
Av.
S.D.
No.
Av.
S.D.
Interlingual
40
0.34
0.71
137
1.15
1.55
37
0.31
0.6
8
0.07
0.25
a) Formal
28
0.24
0.67
119
1
1.33
33
0.28
0.58
8
0.07
0.25
b) Semantic
12
0.1
0.33
18
0,15
0.48
4
0.03
0.22
0
0
0
Intralingual
97
0.82
1.08
135
1.13
1.3
52
0.44
0.9
5
0.04
0.2
a) Formal
25
0.21
0.5
57
0.48
0.89
33
0.28
0.7
2
0.02
0.13
b) Semantic
72
0.61
0.88
78
0.65
1.08
19
0.16
0.41
3
0.02
0.16
The results of our analysis show that the most frequent lexical errors in verbs are intralingual semantic lexical
errors with an average of 0.61 errors per participant. The categories that follow are interlingual formal errors with an
average of 0.24, intralingual formal errors (0.21) and interlingual semantic errors (0.1). As far as the most frequent
lexical errors in nouns are concerned, the analysis reveals that these are interlingual formal errors with an average
of 1 error per participant. The following categories are intralingual semantic errors (0.65), intralingual formal errors
(0.48) and interlingual semantic errors (0.15). The most frequent lexical errors in adjectives are interlingual formal
errors and intralingual formal errors with an average of 0.28 errors, which are followed by intralingual semantic errors
(0.16), and interlingual semantic errors (0.03). Finally, the most frequent lexical errors in adverbs are interlingual
formal (0.07). The categories that follow are intralingual formal errors and intralingual semantic (0.02).
To answer our fourth research question, we classified lexical errors into further categories and subtypes and
examined their evolution across the proficiency levels. Table 5 offers the absolute number and mean figures for
each of the lexical error broad categories at the three proficiency levels for verbs.
Table 5. Evolution of the lexical errors in verbs across levels.
Interlingual
Formals
Intralingual
Semantic
Formal
Semantic
Level of
proficiency
No.
Av.
S.D
No.
Av.
S.D
No.
Av.
S.D
No.
Av.
S.D
B1
9
0.28
0.81
5
0.16
0.37
9
0.28
0.58
24
0.75
0.98
B2
18
0.38
0.79
5
0.11
0.37
8
0.17
0.38
32
0.68
0.84
C1
1
0.03
0.16
2
0.05
0.22
8
0.2
0.56
16
0.4
0.84
The results of our analysis show that the most frequent lexical errors in verbs at all proficiency levels are
intralingual semantic errors. An exploration of their evolution reveals that almost all the categories of lexical errors
in verbs, interlingual (formal and semantic) and intralingual semantic, tend to decrease as the proficiency level in
SFL of the Greek students increases, whereas we can see a slight increase of the intralingual formal errors from
the B2 level to the C1 level.
In addition to this, we have also examined the evolution of the different subtypes of lexical errors as classified
into the taxonomy presented in the method section. The results of this additional and more detailed analysis
show that at the B1 level the most frequent types of lexical error in verbs are errors in the use of ser and estar, the
confusion between lexemes with common semes, adaptations to the FL of L1 or Ln words, and the creation of
non-existent words. Similar results as concerning the most frequent lexical error subtypes were found for the B2
level. Finally, at the C1 level, the most frequent types of lexical errors in verbs were the confusion between lexemes
with common semes, errors in the use of ser and estar, and the creation of non existent words.
RLyLA Vol. 09 (2014), 67-78 | 73
Table 6 shows the absolute number and mean figures for each of the lexical error broad categories
at the three proficiency levels for nouns.
Table 6. Evolution of the lexical errors in nouns.
Interlingual
Formal
Intralingual
Semantic
Formal
Semantic
Level of
proficiency
No.
Av.
S.D
No.
Av.
S.D
No.
Av.
S.D
No.
Av.
S.D
B1
37
1.16
1.22
4
0.12
0.42
13
0.41
0.8
19
0.6
1.21
B2
47
1
1.32
10
0.21
0.59
31
0.66
1.09
39
0.83
1.19
C1
35
0.88
1.44
4
0.1
0.38
13
0.33
0.66
20
0.5
0.78
Regarding the lexical errors produced in nouns, our analysis reveals that the most frequent errors for all
proficiency levels are interlingual formal lexical errors. If we examine the evolution of lexical errors in nouns, the
results of our analysis generally show that they tend to decrease in all the categories of lexical errors, interlingual
(formal and semantic) and intralingual (formal and semantic).
A more detailed analysis of the evolution of different subtypes of lexical errors in nouns shows that, at the B1
level, the most frequent types of lexical error in nouns are adaptations to FL, intralingual errors in gender, and code
switching. In the same way, at the B2 level, adaptations to FL are the most frequent type of lexical error in nouns.
The types of lexical error that follow are intralingual errors in gender and the creation of non-existent words. Finally,
adaptations to FL is again the most frequent type of lexical error at the C1 level followed by confusion between
lexemes with common semes, and the creation of non-existent words, respectively.
Table 7 presents the absolute number and mean figures for each of the lexical error broad categories at the
three proficiency levels for adjectives.
Table 7. Evolution of the lexical errors in adjectives.
Interlingual
Formal
Intralingual
Semantic
Formal
Semantic
Level of
proficiency
No.
Av.
S.D
No.
Av.
S.D
No
Av.
S.D
No.
Av.
S.D
B1
7
0.22
0.49
0
0
0
9
0.28
0.68
3
0.09
0.3
B2
13
0.28
0.58
3
0.06
0.32
16
0.34
0.76
10
0.21
0.46
C1
13
0.32
0.66
1
0.03
0.16
8
0.2
0.69
6
0.15
0.43
Examination of lexical error production in adjectives reveals that the most frequent errors at the B1 and B2
levels are intralingual formal lexical errors. Whereas the most frequent category of lexical error in adjectives at the
C1 level is interlingual formal lexical errors. Additionally, the results of the analysis of the evolution of lexical errors
in adjectives show that interlingual formal lexical errors and intralingual semantic lexical errors increase as the
proficiency level of participants increases. These are the most outstanding findings for this set of results.
The detailed analysis of the evolution of the different types of lexical errors in adjectives reveals that, at the B1
level, the most frequent type of lexical error in adjectives are the confusion between lexemes with common semes,
intralingual errors in gender, and the creation of non-existent words. Regarding the B2 level, the most frequent type
of lexical error in adjectives is the creation of non-existent words. The following most frequent types are changes
in derivatives of the same root, and interlingual errors in gender. At the C1 level, the most frequent types of lexical
error in adjectives are interlingual errors in gender, the creation of nonexistent words, and intralingual errors in
gender. Less frequent lexical error categories are confusion between lexemes with common semes, and changes
in derivatives of the same root.
Finally, Table 8 displays the absolute number and mean figures for each of the lexical error broad categories at
the three proficiency levels for adverbs.
RLyLA Vol. 09 (2014), 67-78 | 74
Table 8. Evolution of the lexical errors in adverbs.
Interlingual
Formal
Intralingual
Semantic
Formal
Semantic
Level of
proficiency
No.
Av.
S.D
No.
Av.
S.D
No.
Av.
S.D
No.
Av.
S.D
B1
2
0.06
0.25
0
0
0
0
0
0
1
0.03
0.18
B2
2
0.04
0.2
0
0
0
0
0
0
0
0
0
C1
4
0.1
0.3
0
0
0
2
0.05
0.22
2
0.05
0.22
Our last set of results concerns lexical error production in adverbs. They show that the most frequent category
of lexical error in adverbs for all proficiency levels is interlingual formal lexical errors. Lexical errors in adverbs have
a very weak presence in our corpus. Interlingual formal errors and intralingual semantic errors in adverbs tend to
increase as the linguistic level of our participants increases but this increase shows an unstable pattern with a
slight decrease at the B2 level, whereas the intralingual formal errors appear only in the C1 level.
Additionally, the analysis of the evolution of the different subtypes of lexical errors in adverbs shows that, at
the B1 level, the most frequent types of lexical error are adaptations to FL and changes in derivatives of the same
root. At the B2 level, the only type of lexical error that appears is the adaptation to FL. Adaptation to FL is the most
frequent subtype of lexical error at the C1 level. These include the creation of non-existent words, errors caused
by circumlocution and errors in collocations.
6. DISCUSSION
Our results show that there is a quantitative difference in the production of lexical errors in the different word
classes. Specifically, lexical errors are most frequent in nouns, followed by verbs, then adjectives and finally
adverbs. These findings answer the first research question. Even when relative measures are taken, we found that
for nouns, which are the most frequent word class, lexical errors are more common than in the remaining classes.
Although figures tend to be similar for adjectives and adverbs at the B2 and C1 proficiency levels.
Previous research has shown (e.g. Kim, McGregor and Thompson, 2000; Bornstein et. al. 2004; Gentner and
Baroditsky, 2009; Li and Fang, 2011) that nouns are the first class of words acquired, and in our data nouns were
the most frequent word class in the output of Greek students studying Spanish as a foreign language. Besides,
lexical errors in nouns present the highest percentage over total words produced of all the word classes. In other
words, not only are nouns the most frequent word class, they also present the highest number of lexical errors in
relative terms. This can be accounted for by problems in mastering nouns which goes well into high proficiency
levels. In conclusion, nouns might be more difficult to command than we had previously assumed. Gender errors
and lexical coinages seem to be responsible for these higher numbers.
Lexical errors in verbs have been found to be numerous as well, but smaller percentages than with nouns
are found. This might reveal that learners have apparently more problems in mastering nouns than verbs. The
multiple content and syntactic information they encode might explain their high cognitive load. Furthermore, verbs
have proved the most difficult word class to be acquired and they require the most repetitions to be mastered or
acquired successfully (San Mateo Valdehíta 2003/2004). Accordingly, assumptions that verbs and the Spanish
verbal system are difficult to command might act as a warning for learners to take more caution when producing
verbs, and hence the smaller percentage of lexical errors than in nouns.
The weak presence of adjectives and adverbs in our corpus cannot be traced back to learners’ good or full
knowledge of them, but rather it might be attributed to avoidance of use (cf. Schachter 1974). It seems that our
participants preferred to use simple syntax in their written output and they avoided adjectives and adverbs, which
are used in long and syntactically more complex phrases and sentences.
Furthermore, the usage based model could also explain the weak presence of adjectives and adverbs in
learners’ production. According to Diessel (2004), who provides a thorough description of this usage based model
of language acquisition and use, the frequency of a linguistic element in the input determines its acquisition and
its appearance in the learners’ output. It seems that this linguistic theory could explain the lack of adjectives and
adverbs in learners’ output. It seems reasonable to think that the input that our participants have received from
their teachers and from the teaching materials is simplified without complex sentences or complicated syntax. If
RLyLA Vol. 09 (2014), 67-78 | 75
we acknowledge that adjectives and adverbs appear in this type of linguistic expression, we can understand the
reason for the weak presence of these word classes in the output of our participants.
Additionally, the results of this study have also revealed different development or evolution trends in the lexical
errors produced in the different word classes. In particular, lexical errors in verbs and nouns tend to decrease as
the proficiency level of the learners increases. On the contrary, lexical errors in adjectives and adverbs show an
increasing tendency.
The decreasing tendency of lexical errors in nouns and verbs can be explained if we focus on the characteristics
of the process of acquisition of SFL. During the first stages of acquisition, Greek learners studying Spanish as
a foreign language seem to use nouns and verbs which they have not yet acquired completely. In other words,
they use nouns and verbs whose formal and semantic aspects have not been internalized at the moment of data
collection. This may explain the considerable amount of lexical errors produced in these word classes, considered
both in absolute and in relative terms. As the proficiency level of learners increases, the formal and semantic
characteristics of nouns and verbs are slowly incorporated in the lexical competence of our participants, and
lexical errors in nouns and in verbs decrease.
In the same line, the development of lexical errors in adjectives and adverbs, can be accounted for in terms
of mastery and use. Again, we are inclined to believe that this reflects real language use. Adjectives and adverbs
become increasingly more frequent in learners’ output and consistently, they are more subject to lexical errors. As
learners’ syntax gets more complicated with the use of adjectives and adverbs, lexical errors start cropping up and
becoming more frequent as well. The type of adjectives and adverbs used might also contribute to this increasing
lexical error production.
According to Schmitt (1998), adjectives and adverbs appear to be learned at later stages of the acquisition
of the FL, so learners begin to use them at advanced stages, when their output becomes more complex. Thus,
errors in adjectives and adverbs will start to come up at later stages and become increasingly more frequent.
Furthermore, taking into account the results by Gentner and Baroditsky (2009) and the studies by Kim, McGregor
and Thompson (2000), Ogura et al. (2006) and Li and Fang (2011) with regard to the importance of the input, it
seems that the input that the learners of the advanced stages receive is richer in adjectives and adverbs. This is
also in line with the usage-based model commented previously. In this sense, we might expect more adjectives
and adverbs in the output of advanced FL learners. Nevertheless, they still present gaps in their knowledge, and
this is the reason for lexical errors in these word classes as well, especially at advanced linguistic levels.
Another interesting result in our study is the different types of lexical errors which are characteristic of the
different word classes. Likewise, these lexical error types in word class are constant across all levels. This points
to word class as a stronger predictor of lexical error type than proficiency level and answers our third and fourth
research questions.
In this sense, the most frequent lexical error type in nouns for all linguistic levels has been found to be formal
lexical errors and especially adaptations to FL. This result can be explained by alluding to the process of the
acquisition of nouns. As previous research has shown (Gentner, 1981; Kim, McGregor and Thompson, 2000;
Bornstein et al. 2004; Gentner and Baroditsky, 2009), nouns are the first word class to be acquired because they
are more transparent in their meaning, since they are used to label concrete objects. Indeed, the results of our
study confirm the early acquisition of the meaning of nouns given the fact that the most frequent lexical errors in
nouns are related to their form.
By contrast, the most frequent lexical error type in verbs for all the linguistic levels are intralingual semantic
lexical errors. Specifically, we observe a strong presence of lexical errors in the use of ser and estar and in lexemes
with common semes. Similarly, the explanation for these results may reside in the process of acquisition of verbs.
According to the studies by researchers such as Gentner (1981) or Caselli et al. (1995), and to the results of the
study of Matanzo Vicens and Reyes Díaz (1998/1999), the semantic characteristics of verbs are more complex
since they indicate actions or states, so the acquisition of their meaning is more difficult than the acquisition of
the meaning of nouns. Our results are consistent with this finding, since we observe that the most frequent lexical
errors in verbs are semantic.
Cognitive models of language acquisition and use predict that learners will have problems with linguistic items
that encode conceptualizations not present in their native languages (Pavlenko 2005). This can account for the
frequent learner errors in the Spanish ser-estar distinction, lacking in Greek L1. The process of L2 acquisition will
bring forth a re-conceptualization of the world, and consequently of the distinction deployed by ser and estar.
The most frequent lexical error type in adjectives for all linguistic levels is intralingual formal lexical errors and
more specifically errors because of the creation of non-existent words. According to Matanzo Vicens and Reyes
RLyLA Vol. 09 (2014), 67-78 | 76
Díaz (1998/1999), the nominal categories of words (nouns, adjectives) present a lower degree of semantic difficulty
than the verbal categories of words. Our results are in the same line with the study of Matanzo Vicens and Reyes
Díaz (1998/1999) given the fact that the highest number of lexical errors in adjectives is located in their form.
Finally, the results of our analysis have shown that the most frequent lexical error type in adverbs are interlingual
formal lexical errors and especially errors because of the adaptations of L1 words to the FL. In this case, it seems
that the formal aspect is the most difficult to be acquired, as well. Nevertheless, the weak presence of lexical
errors in adverbs does not allow to reach an accurate and definite conclusion and further research in this respect
is called for.
7. CONCLUSION
Data of the present study points to the fact that there is a relationship between the type of word and the
production of lexical errors. Furthermore, the evolution of lexical errors presents different trends in the different
word classes. Accordingly, our results lead us to the conclusion that word class is a stronger predictor of the
production of specific lexical error types than proficiency level.
Nevertheless, the lack of studies in the acquisition of the different word classes in FL, has forced us to base the
explanations of some of our results on studies that have investigated the order of acquisition of the word classes in
the mother tongue. Interpretations of results must, therefore, follow with caution and further research is warranted
to further understand the relationship between word class, lexical error production and proficiency level.
Given this lack of studies in FL, we believe that future research is needed in the field of the acquisition of the
different word classes in FL. Linguistic production of learners of Spanish of different mother tongues should be
included in future research in this area in order to find out the parameters of this process.
Acknowledgements
We want to thank Aubrey Wade, MPhil, for having reviewed this version of the paper. We are also thankful to the
anonymous reviewers for providing interesting suggestions to help improve the quality of the paper. Any remaining
errors are our own.
REFERENCES
Agustín Llach, M. and Barreras Gómez, M. (2007). “Childrens’ Characteristics in Vocabulary Acquisition and Use
in the Written Production”, RESLA, 20: 9-26.
Agustín Llach, M., Moreno Espinosa, S. and Fernández Fontecha, A. (2005). “Responding to Different Composition
Topics: A Quantitative Analysis of Lexical Error Production”, Glosas Didácticas, 13: 128-140.
Bornstein, M., Cote, L., Maital, S., Painter, K., Park, S., Pascual, L., Pêcheux, M., Ruel, J., Venuti, P. and Vyt, A. (2004).
“Cross – Linguistic Analysis of Vocabulary in Young Children: Spanish, Dutch, French, Hebrew, Italian, Korean,
and American English”, Child Development, 75/4: 1115-1139. doi:10.1111/j.1467-8624.2004.00729.x
Caselli, M. Ch., Bates, E., Casadio, P., Fenson, J., Fenson, L., Sanderl, L., Weir, J. (1995). “A Cross-Linguistic Study
of Early Lexical Development”, Cognitive Development, 10: 159-199. doi:10.1016/0885-2014(95)90008-X
Corder, S.P. (1967). “The Significance of Learners’ Errors”, IRAL, 5: 161-170. doi:10.1515/iral.1967.5.1-4.161
Corder, S.P. (1973). Introducing applied linguistics. Middlesex: Penguin
David, A. (2008). A developmental perspective on productive lexical knowledge in L2 oral interlanguage. Journal of
French Language Studies, 18/3: 315-331. doi:10.1017/S0959269508003475
Diessel, H. (2004). The acquisition of complex sentences. Cambridge: Cambridge University Press. doi:10.1017/
CBO9780511486531
Fernández, S. (1997). Interlengua y análisis de errores en el aprendizaje del español como lengua extranjera.
Madrid: Edelsa.
Gentner, D. (1981). “Some Interesting Differences Between Verbs and Nouns”, Cognition and Brain Theory, 4/2:
161-178.
Gentner, D. (1982). “Why Nouns are Learned Before Verbs: Linguistic Relativity Versus Natural Partitioning”, in: S.A.
Kuczaj (ed.), Language Development Vol. 2 Language, thought, and Culture. N.J.: Erlbaum. pp. 301-334.
Gentner, D. and Baroditsky, L. (2009). “Early Acquisition of Nouns and Verbs. Evidence from Navajo”, in: V.
Gathercole (ed.), Routes to language: Studies in honour of Melissa Bowerman. New York: Taylor & Francis.
pp. 5-36.
RLyLA Vol. 09 (2014), 67-78 | 77
Jarvis, S. (2000). “Methodological Rigor in the Study of Transfer: Identifying L1 Influence in the Interlanguage
Lexicon”, Language Learning, 50/2: 245-309. doi:10.1111/0023-8333.00118
Jodar, R. (2006). Análisis de errores léxicos, morfosintácticos y gráficos en la lengua escrita de los aprendices polacos
de español. PhD. University Adam Mickiewitz of Poznan, en: http://www.mecd.gob.es/dctm/redele/MaterialRedEle/Biblioteca/2007_BV_08/2007_BV_08_10RaulFernandez.pdf?documentId=0901e72b80e5384f
[Access date: 26/12/2013]
Housen, A., Pierrard M., Van Daele S., Bulté, B. (2008). Investigating lexical proficiency development over time? the
case of Dutch-speaking learners of French in Brussels. Journal of French Language Studies, 18/3: 277–298.
Kim, M., McGregor, K. and Thomson, C. (2000). “Early Lexical Development in English-and Korean-Speaking
Children: Language - General and Language-Specific Patterns”, Journal of Child Language, 27: 225-254.
doi:10.1017/S0305000900004104
Lennon, P. (1991). Error and the very advanced learner. IRAL 29/1: 31-43. doi:10.1515/iral.1991.29.1.31
Li, H. and Fang, A. (2011). “Word Frequency of the CHILDES corpus: Another perspective of child language
features”, ICAME Journal, 35: 95-116.
Matanzo Vicens, G. and Reyes Díaz, M.J. (1998/1999). “Adquisición de las Categorías Gramaticales por Alumnos
de Enseñanza Universitaria y Preuniversitaria (Bachillerato)”, Philologica Canariensia: Revista de filología de
la Universidad de las Palmas de Gran Canaria, 4/5: 187-196.
Myles, F. (2004). “From Data to Theory: The Over-Representation of Linguistic Knowledge in SLA”, Transactions of
the Philological Society, 102/2: 139-168. doi:10.1111/j.0079-1636.2004.00133.x
Naigles, L. and Hoff-Ginsberg, E. (1998). “Why are some verbs learned before other verbs? Effects of input
frequency and structure on children’s early verb use”, Journal of Child Language, 25: 95-120. doi:10.1017/
S0305000997003358
Pavlenko, A. (2005). “Bilingualism and Thought” in Kroll, J. F., and de Groot, A. M. B. (eds). Handbook of
Bilingualism:Psycholinguistic Approaches. Oxford University Press, New York. pp. 433-453.
Ogura, T., Dale, Ph., Yamashita, Y., Murase, T. and Mahieu, A. (2006). “The Use of Nouns and Verbs by Japanese
Children and their Caregivers in Book-Reading and Toy-Playing Contexts:, Journal of Child Language, 33/1:
1-29. doi:10.1017/S0305000905007270
Palapanidi, K. (2012). Los errores léxicos de los aprendientes griegos de español. Datos de un corpus de producción
escrita. Saarbrücken: Editorial Académica Española.
Robinson, P. and Ellis, N.C. (2008). Handbook of Cognitive Linguistics and Second Language Acquisition. London:
Routledge.
Schachter, J. (1974). “An error in error analysis”, in Language Learning, 24: 205-214. doi:10.1111/j.1467-1770.1974.
tb00502.x
Schmitt, N. (1998). “Tracking the Incremental Acquisition of Second Language Vocabulary: a Longitudinal Study”,
Language Learning, 48/2: 281-317. doi:10.1111/1467-9922.00042
Šifrar Kalan, M. (2009). “Disponibilidad léxica en español como lengua extranjera: el cotejo de las investigaciones
en Eslovenia, Salamanca y Finlandia”, Verba Hispánica, 17: 165-182.
Sökmen, A. J. (1997). “Current trends in teaching second language vocabulary”, in: N. Schmitt and M. McCarthy
(eds.), Vocabulary: Description, Acquisition and Pedagogy. Cambridge: Cambridge University Press. pp.
237-257.
Tardif, T., Shatz, M. and Naigles, L. (1997). “Caregiver Speech and Children’s Use of Nouns Versus Verbs: A
Comparison of English, Italian and Mandarin”, Journal of Child Language, 24/3: 535-565. doi:10.1017/
S030500099700319X
VanPatten, B. (2004). Processing instruction: theory, research, and commentary. New Jersey: Lawrence Erlbaum
Associates.
Vázquez, G. (1991). Análisis de errores y aprendizaje de español/lengua extranjera. Frankfurt: Peter Lang.
Wray, A. and Fitzpatrick, T. (2010). “Pushing learners to the extreme: The artificial use of prefabricated material in
conversation”, Innovation in Language Learning and Teaching, 4: 37-51. doi:10.1080/17501220802596413
Zimmerman, R. (1987). “Form-Oriented and Content-Oriented Lexical Errors in L2 Learners”, IRAL, 25: 55-67.
RLyLA Vol. 09 (2014), 67-78 | 78
Vol. 9 año 2014, 79-87
EISSN 1886-6298
Inventario fraseológico de las groserías en estudiantes de San Luis Potosí
Marco Antonio Pérez Durán
Oscar Arriaga Olguín
Universidad Autónoma de San Luis Potosí (México)
Resumen: Este trabajo se inserta en el campo de la lexicografía y su objetivo es hacer la descripción formal de unidades
fraseológicas sobre groserías rastreadas en un corpus de estudiantes de San Luis Potosí, México, a partir de la elaboración
del inventario fraseológico y su clasificación en nominales, adjetivales, verbales y adverbiales, de acuerdo con la muestra
obtenida. El método utilizado para llevar a cabo este trabajo consistió en aplicar una encuesta a 300 informantes (150 mujeres
y 150 hombres) de primaria, secundaria y preparatoria (50 mujeres y 50 hombres por cada subsistema educativo) de la capital
potosina.
Palabras clave: unidades fraseológicas, groserías, lexicografía.
Abstract: This study belongs to the field lexicography and its goal is to make the formal description of phraseological units in a
corpus of curses collected among schoolchildren in the city of San Luis Potosí, México. It is based on an inventory of phrases
classified as nominal, adjectival, verbal or adverbial. The method used in the study consisted of interviews with 300 respondents,
50 male and 50 female in each of the three age subdivisions, of the Mexican school system (primaria, secundaria, preparatoria).
Keywords: phraseological units, lexicography, to say grossness or to say bad words.
1. Introducción
Este trabajo se inscribe en el campo de la lexicografía, disciplina que estudia el estrato léxico del lenguaje e
interesa para los fines de este artículo en el análisis de las unidades fraseológicas (UFs) del español, referidas
a palabras tabú (groserías), frases prefabricadas de las que disponen los hablantes de nuestra lengua para
referirse con otras palabras a una misma situación: ej. bájale de huevos puede ser remplazado por tranquilizarse
o calmarse. Estas expresiones que aparecen en el lenguaje oral en múltiples situaciones comunicativas forman
parte de factores extralingüísticos que se ven reflejados en el inventario fraseológico de un determinado grupo
social; en este caso, jóvenes estudiantes de primaria, secundaria y preparatoria de San Luis Potosí. De acuerdo
con Álvarez Martínez, (2001: 535), estas expresiones son consideradas como palabras tabú y pertenecen a las
hablas vulgares.
El objetivo de nuestro trabajo es hacer la descripción formal de UFS sobre groserías rastreadas en un corpus
de estudiantes, a partir de la elaboración del inventario fraseológico y su clasificación en nominales, adjetivales,
verbales y adverbiales, de acuerdo con la muestra obtenida. El método utilizado consistió en la aplicación de una
encuesta a 300 informantes de primaria, secundaria y preparatoria (50 mujeres y 50 hombres por cada subsistema
educativo) de la capital potosina a partir de la instrucción Escriba todas las groserías que conozca1.
Los informantes procedieron al llenado de la encuesta durante los tres minutos otorgados, periodo suficiente
para que accedieran a la información almacenada en su mente y pudieran contestar correctamente lo que se les
pidió. Una vez obtenida la muestra, se procedió a la revisión, corrección ortográfica, homologación y selección de
1
En este estudio no se considera el análisis sociolingüístico por cada grupo de informantes, que será parte de otra investigación.
| 79
Marco Antonio Pérez Durán y Oscar Arriaga Olguín
las frases para el artículo2; para ello, se utilizó el programa de cómputo Excel, que ayudó en la organización de la
base de datos. Se obtuvieron 152 frases que se analizaron como corpus y se clasificaron (tomando en cuenta el
criterio de Coseriu, 1977) en cuatro grupos: nominales, adjetivales, verbales y adverbiales.
2. Marco teórico
Las manifestaciones peyorativas que aparecen en la lengua hacen referencia a maldiciones que están
influenciadas directamente por la comunidad lingüística. Éstas reciben el nombre de palabras tabú, que hacen
referencia a temas desagradables, vulgares o místicos. De acuerdo con Rotaexe (1990: 67), la palabra tabú es de
conducta social y no un mero medio de comunicación, sino de transmisión de significados sociales, ej. pinche
joto por homosexual. Estos significados sociales permiten comprender la relación entre la lengua y la comunidad
como tal.
La frase es una unidad de carácter morfológico, sintáctico o semántico que resulta de la combinación de dos
o más palabras para crear y designar un nuevo significado alejado del que tienen separadamente, es decir, son
dos términos estables en el sistema que forman una unidad semántica compleja y no la suma ni la adición de
sus significados particulares. El proceso de fijación es una particularidad y una característica de las frases que
aparecen cuando son repetidas a lo largo del tiempo y por periodos elevados, de esta manera son contextualizadas
para referirse a una situación específica por parte de un hablante y del grupo social. En palabras de Corpas Pastor
(1997:66), es posible reconocer este tipo de frases, a través de diferentes huellas que imprimen: el orden estable
de los elementos integrantes, la estabilidad en las categorías gramaticales, la imposibilidad de suprimir, insertar o
sustituir un componente y el grado de idiomaticidad.
Gracias al proceso de lexicalización, un sintagma llega a funcionar como una unidad léxica; esto se debe a su
frecuencia fija de uso que establece la comunidad lingüística y permite sea reconocida e integrada totalmente como
una frase, independiente del término palabra. De acuerdo con Fuenzalida (2001:83), es una etapa de transición
desde que inicia el acercamiento mutuo de dos o más vocablos hasta que se unen con vínculo indestructible,
sacrificando alguno de ellos. La frecuencia para la consolidación de la misma frase coincide muchas veces con el
dominio, estratificación y uso que se le da como unidad indivisible, es el caso de las Ufs que forman el centro de
la fraseología; aunque muchas de las Ufs tratadas en este trabajo no cumplen este criterio y se encuentran más
bien en la periferia de la fraseología porque están motivados metafóricamente.
La frecuencia y la disponibilidad son factores característicos y necesarios para determinar el grado de
aceptación de una unidad como frase. De esta manera muchas de las variaciones fraseológicas que se generan a
partir de una no necesariamente formarán parte del inventario léxico reconocido por la comunidad lingüística, ej.
La concha de tu madre, no tendrá mucha aceptación fuera de su contexto, a diferencia de Hijo de tu puta madre,
cuya referencia es más conocida por diferentes comunidades lingüísticas. Desde un punto de vista funcional,
las frases se pueden clasificar en dos tipos: significantes y conexivas, de acuerdo con Coseriu (1977). Las
primeras poseen significado léxico y pertenecen las nominales, adjetivales, verbales, adverbiales, pronominales
y exclamativas, en tanto que las conexivas son aquellas que poseen significado categorial y forman parte las
conjuntivas y las prepositivas.
3. Análisis cuantitativo
Del corpus obtenido se registraron 152 frases, distribuidas en 50 frases nominales que representan 32,87%,
31 frases adjetivales que ascienden a 20,39%, 68 frases verbales que conforman 44,73% y tres frases adverbiales
que constituyen 2% de la totalidad de la muestra. A continuación se presentan todas las frases por frecuencia de
aparición de la muestra y clasificadas en nominales, adjetivales, verbales y preposicionales3.
3.1 Frases nominales
Las groserías en México son estructuras fraseológicas altamente productivas y funcionales en la lengua
hablada; éstas permiten generar un sinfín de nuevas unidades para referirse al insulto. Se presenta el cuadro de
frases nominales por frecuencia de aparición de la muestra.
2
3
En este tipo de trabajos se obtienen nominales, adjetivales, verbos, frases, etcétera, pero para el objetivo principal se examinarán únicamente las frases.
La muestra clasificada que se presenta en este trabajo forma parte de la base de datos sobre disponibilidad léxica de las groserías en México que en la actualidad nos
encontramos realizando.
RLyLA Vol. 09 (2014), 79-87 | 80
animal rastrero
gata culera
hija de mierda
hija de perra
hija de puta
hija de tu madre
hijo de mierda
hijo de perra
hijo de puta
la concha de tu madre
perro desgraciado
rata inmunda
tetas asesinas
verga chingada
puta madre
tlacuache bodeguero
cara de huevo
cara de nalga
cara de pene
cara de perra
cara de perro
cara de picha
cara de pito
cara de vagina
hija de tu madre
las bolas
mis polainas
ni madres
negro asqueroso
perra madre
perra malparida
cabeza de nalga
cabeza de huevo
cabeza de pija
cabeza de verga
cara de caca
cara de culo
hija de tu pinche madre
hijo de la chingada
hijo de la verga
hijo de tu perra madre
hijo de tu puta madre
hijo del coño
meco mal aventado
pedazo de cadáver
pedazo de feto
pedazo de mierda
tu puta madre
tu reputa madre
verga de burro
De acuerdo con Reeves, Hirsh y Golinkoff (1999: 176), en cada selección de vocablos para la construcción
de unidades fraseológicas hay una tarea de decisión léxica que consiste en considerar todas las posibles
combinaciones para generar nuevas unidades. A su vez, éstas deberán contar necesariamente con parecidos
referenciales que sirvan de base para la construcción de la unidad. Por tal motivo, en el estudio fraseológico
de las groserías, muchas de ellas tienen como base contextos de familiaridad, ej. hijo(a), madre(s); en otras
ocasiones utilizan partes del cuerpo como soportes semánticos o referentes, ej. cara, verga, tetas, cabeza; otras
utilizan nombre de animales para generar el insulto, ej. perra, gata, tlacuache, rata; o bien, marcan profesión o
cualidad, ej. pedazo, puta, etcétera. La base del contexto de familiaridad va acompañada de la preposición de
que marca direccionalidad y especificidad en la estructuración del insulto, ej. hija de, hijo de, cara de, etcétera, y
los determinativos (en el complemento) se refieren a elementos escatológicos, ej. mierda, caca; partes del cuerpo,
ej. nalga, pene, picha (vagina), pito (pene); oficios, ej. puta; o bien, con adjetivos determinativos de cualidad, ej.
estúpida, inmunda, entre otros. De acuerdo con lo anterior, se clasifican las frases en:
1. <<N + N>>
Perra madre
Perra malparida
Puta madre
Tlacuache bodeguero
2. <<Det.(art) + N / Det.(Adj.) + N>>
Las bolas
Mis polainas
3. <<N + A >>
Animal rastrero
Rata inmunda
Negro asqueroso
Tetas asesinas
Verga chingada
Perro desgraciado
Gata culera
4. N+ SP
a. <<N + de + N>>4
Cabeza de huevo
Cabeza de pija
Cabeza de verga
Cara de caca
Cara de culo
Pedazo de cadáver
Pedazo de feto
Pedazo de mierda
4
Cara de huevo
Cara de nalga
Cara de pene
Cara de perra
Cara de perro
Cara de picha
Cara de pito
Cara de vagina
Hija de puta
Hija de perra
Hija de mierda
Hijo de mierda
Hijo de perra
Hijo de puta
Verga de burro
La presente fórmula propuesta por García-Page (2008), destaca las cualidades de la preposición, colocándola como eje prepositivo en gran parte de la muestra. Las
atribuciones que en ella hay pueden ser establecidas en la fórmula <<Nominal + Prep. + Nominal>>. Al colocarla en el mismo continuum informativo, el pleonasmo
formulario respondería a la misma frase escrita.
RLyLA Vol. 09 (2014), 79-87 | 81
b. <<N + Prep. + Pron. +N>>
Hija de tu madre
c. <<N + Prep. + Pron. + N + N>>
Hijo de tu perra madre
Hijo de tu puta madre
d. <<N + Prep. + Pron. + Det.(Adj) + N >>
Hija de tu pinche madre
e. <<N + Prep. + Det.(Art.) + N>>
Hijo del coño
Hijo de la chingada
Hijo de la verga
f.
<<N + A + A>>
Meco mal aventado
g. <<Det.(art) + N + Prep. + Pron. + N>>
La concha de tu madre
h. <<Conj.(adversativa) + N>>
Ni madres
i.
<<A(posesivo)+ N + N>>
Tu reputa madre
Tu puta madre
Toda frase referida a grosería adquiere mayor presencia cuando en ella se relacionan dos palabras que competen
a diferentes entidades referenciales. Por ejemplo, los vocablos hijo y mamá pertenecen a un ámbito netamente
familiar, y se encuentran asociados a valores de amistad, lealtad, respeto y cariño. Cuando son utilizados en el
ámbito de la grosería adquieren otros valores que enfatizan la pertenencia a la familia, pero de manera agresiva
o negativa. De acuerdo con Moreno Fernández (2008: 136), los componentes de la norma sociolingüística que
regulan y registran los usos particulares para cada una de las frases están clasificados por su variabilidad, es decir,
las groserías con términos de vocablos de familia están más cercanas a la lengua coloquial que aquellas otras
expresiones cuyos vocablos no necesariamente pertenecen al ámbito familiar. Por ello, muchas de las groserías
tienen o toman de base el vocablo de familia que están asociadas a complementos de orden escatológico, ej.
mierda, a complementos de orden metafórico, ej. perra, o bien, a complementos de oficio, ej. puta.
3.2 Frases adjetivales
Las frases adjetivas se definen como una estructura morfológica de composición entre dos unidades, una
que sirve de núcleo —en este caso un adjetivo— más un determinante cualquiera —otro adjetivo—, ej. pinche
estúpida5, puta estúpida. Según García-Page Sánchez (2008), este tipo de frases tiene un inventario muy reducido
en la lengua, por lo que su uso no es tan frecuente como la nominal o la verbal. A continuación se presentan las
frases adjetivales que aparecieron en la muestra en el siguiente cuadro:
bueno para nada
chingada madre
culero de mierda
pinche bato
pinche cabeza de mojón
pinche joto
pinche madre
pinche pendejo
pinche perra
pinche perro
pinche prostituta
5
ojete estúpido
pinche bastardo
pinche cabeza de mojón
pinche cabrón
pinche cola de zorra
pinche estúpida
pinche estúpido
pinche idiota
pinche joto de mierda
pinche vergas
pinches chingaderas
pinche estúpida malparida
pinche golfa
pinche lelo
pinche mamón
pinche puto
puta estúpida
puta arrabalera
su pinche madre
tu pinche madre
De acuerdo con el diccionario de la Real Academia de la Lengua Española (2001), la voz pinche tiene dos acepciones, una referida a las personas que prestan servicio
en la cocina, cuya clasificación es un nominal y otra relacionada con un término despectivo que se utiliza en México cuyo significado hace referencia a despreciable, esta
última definición es un adjetivo y es la que se utiliza en este trabajo.
RLyLA Vol. 09 (2014), 79-87 | 82
En la elaboración de la grosería se puede decir que hay una serie de principios sociolingüísticos que ronda la
estructuración de la frase; los hablantes eligen la grosería considerando aspectos psicosociológicos (actitud del
hablante, situación psicológica y actitud comunicativa), que son factores determinantes para la revitalización de
la grosería como tal. Obsérvese la diferencia que hay entre los ejemplos a y b.
a. pinche perra
b. bueno para nada
En el primer ejemplo, la frase del tipo A + N es una construcción metafórica cuyo contenido referencial está
ligado al sustantivo perra, es decir, el hablante añade ciertas características del animal que las proyecta cuando
las enuncia y se intensifican cuando agrega el adjetivo pinche, así la frase pinche perra será usada para referirse
a una mujer desgraciada, ojeta, culera, maldita, etcétera. A diferencia del segundo ejemplo, en donde la frase
A + Prep. + N no necesariamente marca o intensifica la acción peyorativa, bueno para nada es una expresión más
coloquial para referirse a objetos que no cumplen con la función para la que fueron hechos. Una frase sirve más
para insultar a una persona que la otra, porque está más asociada a objetos de mala calidad. A continuación se
hace la clasificación de este tipo de frases.
1. << A + Prep. + N>>
Culero de mierda
Bueno para nada
2. <<A + N + Prep. + N>>
Pinche cabeza de mojón
Pinche cola de zorra
3. <<A + A + N>>
Pinche estúpida malparida
4. <<A(posesión) + A + N(sing.)>>
Tu pinche madre
5. <<A + N(sing.)>>
Chingada madre
Pinche bato
Pinche joto
Pinche madre
Pinche prostituta
Pinche perra
Pinche perro
Pinche pendejo
6. <<A(sing/plural) + N(plural)>>
Pinche vergas
Pinches chingaderas
7. <<Det.(adj.) + A>>
Pinche bastardo
Pinche cabrón
Pinche estúpido
Pinche estúpida
Pinche idiota
Ojete estúpido
Pinche lelo
Pinche mamón
Pinche puto
Negro asqueroso
Pinche golfa
8. <<A + N + Prep. + N>>
Pinche joto de mierda
Pinche cabeza de mojón
9. <<A + A + N>>
Su pinche madre
RLyLA Vol. 09 (2014), 79-87 | 83
10. <<A + A >>
Puta arrabalera
Puta estúpida
3.3 Frases verbales
Otra clase de fraseologísmos es aquella de estructura verbo + complemento. Estas frases del tipo verbales
son las más abundantes en el español, ej. chingas a tu puta madre significa molesta a tu mamá. De acuerdo
con Moreno Fernández (2008), los matices significativos que se dan a algunas frases son establecidos por la
misma comunidad lingüística, y las nuevas unidades significativas que son utilizadas para enunciar la misma
realidad están condicionadas necesariamente por esa misma comunidad. Ésta es la razón por la que de una frase
cualquiera hay variaciones léxicas que pueden tener la misma base verbal y diferente complemento o viceversa.
A continuación se enuncian las frases verbales de la muestra.
agárrame ésta
cállate el hocico
chinga tu cola
chinga tu madre
chinga tu perra madre
chingas a tu puta madre
chúpame
chúpamela
jálamela
jálatela
me la mamas
me la pelas
no chingues huevos
no estés mamando
no seas joto
no seas pendejo
no vales madre
no vales verga
pícate el culo
pícate la cola
rósame el coño
siéntamelo
te la pelaste
chinga tu puta, perra, bomba madre
chinga tu vida
chupa la cabezona
chúpamelo
chúpatelo
concha tu madre6
mámamela
me la mamas
me la pelas
no chingues
no jodas
no mames
no manches
pícatelo
te la pelaste
vale madre
vale verga
vete a la mierda
vete a la verga
vete a la víctor
vete al coño
vete al culo
bájale de huevos
bájale de ovarios
cagando el palo
chíngate
chúpala
chúpala pichón
chúpalo
chúpame el meón
chúpame el pito
chúpame ésta
chúpame la verga
chúpame los huevos
chúpate el culo
jódete
me cagas
no estés chingando
no te pases de verga
vales verga
vete a la verga, maricón
vete al diablo
vete al huevo
vete al pito
vete al coño
Los informantes de San Luis Potosí utilizan agarrar, chingar, chupar, jalar, mamar, valer, ser, picar, joder,
cagar, pelar, sentir e ir como verbos característicos para la formación de groserías y la mayoría tiene que ver con
asociaciones sexuales. De ellos, mamar, chingar, cagar, sin especificativo constituyen ya una grosería. Los otros
verbos necesitan de su complemento para referirse a una frase peyorativa. Ser grosero es una conducta activa en
la que la formación sintáctica está presente a través de cierto tipo de verbos, en este caso acción, conductuales
y desplazamiento. En los verbos que por sí solos tienen marca peyorativa, el complemento funciona como un
intensificador, ej. me cagas los huevos, aquí el complemento huevos refiere a los testículos.
Otro ejemplo se aprecia con la frase mámamela, cuya referencia se hace a la acción de succionar algo por
alguien, en este caso el órgano masculino. Estos verbos no necesariamente requerirán del complemento como
agarrar, que no proyecta referencialmente la acción sexual, ej. agárrame ésta, que refiere su complemento al pene.
Otro ejemplo de verbos que requieren del complemento para marcar término peyorativo es ir en vete a la mierda,
que significa ‘no estar molestando’. El nominal mierda, el pronombre me ni el verbo por sí solos marcan esa
referencia, es sólo a través de una unidad fraseológica.
6
La frase concha tu madre tiene una clase de verbo polisémico que remite a chingar. Posiblemente sea parte de un dialectalismo porque solamente apareció en una
ocasión en la muestra general.
RLyLA Vol. 09 (2014), 79-87 | 84
1. <<V + CD>>
a. <<V + Det.(art.) + N >>
Cállate el hocico
Chupa la cabezona
b. <<V + Pron. + N>>
Chinga tu cola
Chinga tu madre
Chinga tu vida
Concha tu madre
c. <<V + N>>
Vale madre
Vale verga
Vales verga
Agárrame ésta
d. <<V + Pron. + N + N>>
Chinga tu perra madre
e. <<V + Det.(Art.) + N + N>>
Chingas a tu puta madre
f.
<<V + Pron. + N + N + N + N>>
Chinga tu puta, perra, bomba madre
2. <<V + CI + CR>>
Vete a la mierda
Vete a la verga
Vete a la víctor
Vete al coño
Vete al culo
Bájale de ovarios
Vete al diablo
Vete al huevo
Vete al pito
Vete a la verga, maricón
Bájale de huevos
3. <<V + CD + CI / V + CI + CD>>
Chúpala pichón
Chúpame el meón
Chúpame el pito
Chúpame ésta
Chúpame la verga
Chúpame los huevos
Chúpate el culo
Pícate el culo
Pícate la cola
Rósame el coño
4. <<CI + CD + V>>
Me la mamas
Me la pelas
Te la pelaste
5. <<Locución verbal de polaridad negativa>>7
a. <<Adv.(negación) + Estar + gerundio>>
No estés mamando
No estés chingando
b. <<Adv.(negación) + V + N(singular/plural)>>
No vales madre
No vales verga
No chingues huevos
No seas pendejo
No seas joto
7
Todas las oraciones verbales son de polaridad negativa. La descripción gramatical que se desarrolla permite en un primer momento englobar ciertas locuciones que
comparten rasgos entre sí. El segundo punto es que, además el análisis práctico de las locuciones, las estructuras pudieran funcionar como una plataforma en la descripción de groserías.
RLyLA Vol. 09 (2014), 79-87 | 85
c. <<Adv.(negación) + V(plural)>>
No mames
No manches
No jodas
No chingues
d. <<Adv.(negación) + Pron. + V + Prep. + N>>
No te pases de verga
6. <<CI + V>>
Me cagas
7. <<Gerundio + Det. (art.) + N>>
Cagando el palo
8. Verbos cuyo sujeto pronominal opcional o tácito no está fijado en una persona determinada.
a. <<V + Pronominal(acusativo)>>
Chíngate
Chúpala
Chúpalo
Jódete
Chúpame
Siéntemelo
b. <<Verbo + Pronominal(Acusativo) + Pronominal(Dativo)>>
Chúpamela
Jálamela
Jálatela
Chúpamelo
Chúpatelo
Mámamela
Pícatelo
3.4 Frases prepositivas
De acuerdo con García-Page (2008), el inventario de locuciones adverbiales es junto con el de las verbales el
más numeroso; sin embargo, son realmente pocas las locuciones que pueden llamarse adverbiales en virtud de la
estructura de sintagma adverbial que adoptan. En consecuencia, una frase adverbial para este trabajo tendrá la
estructura de un sintagma preposicional. Se muestran las frases adverbiales que aparecieron en el corpus:
a la mierda
a la verga
por la punta
Este grupo está compuesto por tres frases, de las cuales dos están constituidas por preposición a, que
marca direccionalidad o locación, mientras que la preposición por marca agencialidad. De estas frases, la que no
especifica el significado peyorativo es por la punta; no tiene referencia clara. Las otras dos frases son peyorativas
por los nominales mierda y verga. La estructura que prevalece en las adverbiales es:
1. <<Prep + Det. (art) + N (fem sing)>>
Por la punta
A la mierda
A la verga
4. Conclusiones
De acuerdo con lo analizado las frases verbales ocupan la primera posición (68), le sigue las nominales con 50
frases, las adjetivales con 31 ocupan la tercera posición y en el último lugar se ubican las preposicionales con 3.
En ese sentido la composición de la grosería en esta variante dialectal de estudiantes potosinos está relacionada
mas a la estructura verbal y nominal que la adjetival y preposicional. En las nominales abundan los términos
RLyLA Vol. 09 (2014), 79-87 | 86
de familia para generar la frase peyorativa. Vocablos como hija(o), madre van unidos a términos de vocablos
escatológicos (mierda, caca, etcétera), o referentes a los órganos sexuales, (pene, verga, vagina, coño, entre
otros). Se ha observado que la constitución del término familiar más un término despectivo puede formar frases
de este tipo. También dentro del grupo de nominales se demostró que usos metafóricos de animales (perro, perra,
tlacuache) forman parte de la estructura fraseológica.
Los verbos que integran las frases verbales son 13, que pueden formar o estructurar frases de este tipo y de
contenido peyorativo. Sólo tienen contenido sexual expreso: mamar, chingar, joder y contenido escatológico cagar,
es decir, no necesitan de un complemento para referirse a una grosería; los otros sí requieren de ese complemento
para especificar su contenido peyorativo. Los complementos tienen que ver con términos escatológicos (mierda)
o con términos relacionados con los órganos sexuales (pito, verga, etcétera). Piche es el vocablo característico
de la frase adjetiva; es una palabra muy utilizada para referirse a la grosería, por lo menos en esta muestra.
Finalmente, son pocas las frases prepositivas que aparecieron en la muestra y de este grupo solamente dos de
ellas hacen alusión a la grosería.
REFERENCES
Álvarez Martínez, M.A. (2001). “Vulgarismos y Neologismos”, en Alvar, M. (ed.), Introducción a la lingüística
española. Barcelona: Ariel, pp. 533-545.
Corpas Pastor, G. (1997). Manual de fraseología española. Madrid: Gredos.
Coseriu, E. (1977). Principios de semántica estructural. Madrid: Gredos.
Fuenzalida, M. (2001). “Precisiones para un desarrollo teórico-descriptivo de las solidaridades léxicas”. Boletín de
Filología de la Universidad de Chile, Núm. 38: pp. 67-109.
García-Page Sánchez, M. (2008). Introducción a la fraseología española: estudio de las locuciones. Madrid:
Anthropos.
Moreno Fernández, F. (2008). Principios de sociolingüística y sociología del lenguaje. Barcelona: Ariel, Letras.
Real Academia Española. (2001). Diccionario de la lengua española. 22ª ed. Madrid: Espasa.
Reeves, L., Hirsh Pasek, K., y Golinkoff, R. (1999). “Palabras y significado: de los elementos simples a la
organización compleja”, en Berko Gleason, J. y N. Bernstein Ratner (ed.), Psicolingüística. México: McGraw
Hill, pp. 169-225.
Rotaexe, K. (1990). Sociolingüística. Madrid: Síntesis.
RLyLA Vol. 09 (2014), 79-87 | 87
Vol. 9 año 2014, 88-96
EISSN 1886-6298
The use of clickers to assess knowledge in foreign language classes and
their failure to increase reading compliance
Butler University, Indianapolis, United States
Abstract: This is the first quantitative research on reading compliance in FL courses. It investigated the effect of clickers on
learning gains for regularly assigned readings, determined by 16 quiz grades during a semester. 38 intermediate L2 Spanish
students assigned to two group conditions also completed a questionnaire at the end of the semester about their preparedness
for the quizzes and their opinions about the use of clickers. Results indicated that participants in the Clicker condition obtained
significantly lower grades in the quizzes than those in the Paper and Pencil one, despite clickers receiving positive feedback and
comments, and even though students in the Clicker condition reported preparing for class more often than those in the Paper
and Pencil condition.
Keywords: reading compliance, Spanish, foreign language, quiz, clickers.
1. introduction
Most foreign language (FL) students attending college nowadays were born approximately between the early
80’s and 2000. Howe and Strauss (2000) referred to them as the Millennial Generation. They are all “native speakers
of the digital language of computers, video games and the Internet” (Prensky, 2012: 69) rather than becoming
fascinated by and adopting many or most aspects of the new technology at some later point in their lives. That is
the fundamental observation that differentiates what Prensky (2012) calls the Digital Natives (i.e., the Millennials)
from the Digital Immigrants (i.e., all previous generations).
Besides an information technology mindset, Millennial students seem to have zero tolerance for delays
and demand immediate feedback and gratification. According to Spodark (2010), this need for immediacy has
implications in the way they approach reading, i.e., they avoid reading any text of considerable length, even if it is
work-related material. And this picture seems to be no different in the school context.
1.1. Reading compliance at college
In 2000, Burchfield and Sappington reported a steady and dramatic decline on compliance with the required
reading assignments from introductory to graduate-level psychology courses from 1981 to 1997. Reading
compliance was determined by the grade on the first surprise quiz of the semester on an assigned reading. They
found that, on average, only about a third of the students will have completed their reading assignment on any
given day. In 2002, Sikorski, et al. corroborated this trend after surveying 1,178 students at two different institutions
who reported either not reading or reading sparingly the text for their Introductory Psychology class. Similarly, in
2004 Clump, et al. surveyed students from eight psychology courses and they also found that, on average, less
than a third of the students completed the assigned readings before class.
Failing to do the reading assignment and coming to class unprepared are examples of what Burroughs, et al.
(1989) called “destructive resistance” which, not surprisingly, affects students’ learning and achievement. For
example, Karp and Yoels (1976) found that more than three quarters of students from a wide range of departments,
chose not to talk in class, remaining silent and uninvolved, because they have not done the assigned reading.
| 88
The use of clickers to assess knowledge in foreign language classes and their failure to increase reading compliance
In FL classes, remaining silent as a result of not coming prepared to class has greater dramatic consequences.
In the field of second language acquisition (SLA) it is now generally agreed that output, or the language that
learners produce in speech or in writing, plays a central role in the learning process in several ways: a) to test
learners’ developing hypotheses (Swain, 1995), b) to enable learners to search for additional confirmatory or
non-confirmatory evidence after receiving feedback or negative evidence (Gass, 1997), c) to develop fluency and
automatic processing via consistent and successful mapping of grammar to output (Gass and Selinker, 2008), and
d) to force the learner to move from semantic processing to syntactic processing (Swain, 1985).
To the best of my knowledge, there is no study addressing reading compliance in SLA and how it compares to
the rates found in other disciplines. The amount of content to read for FL courses is generally limited to less than
three pages of grammar explanations and examples or lists of vocabulary items. But if FL teachers want to use a
true communicative teaching approach, they have to rely on the students’ preparation for each class to practice
the language in a more natural way. Otherwise, teachers will be forced to introduce and explain the content from
the assigned grammar or vocabulary section before practicing the language, valuable time which could have been
better employed to provide opportunities for practicing and using the language.
One of the most effective ways to promote reading compliance at college is frequent quizzing. There is a
body of research showing that when students are tested on the contents of the assigned readings, with a grade
percentage of their final grade stemming from these quizzes, reading compliance increases. Thorne (2000) noted
that administering randomly assigned quizzes promoted pre-class preparation and increased class attendance.
Similarly, Connor-Greene (2000) found that 60% of her Women and Psychology students, who had daily quizzes,
always completed their reading by the assigned date, with 92% of them reading it before class. Ruscio (2001) data
from four psychology courses indicated that when students were given quizzes at random on less than half of all
class days, they completed the assigned reading at impressive rates (passing an average of 74% of the quizzes
with full credit) and that most students (85.7%) recognized finishing at least half of the assigned readings before
class. Sappington, et al. (2002) found a significant correlation between students’ first surprise quiz grade and final
exam scores and, accordingly, they recommended surprise quizzes and their percentages. Clump, et al. (2004)
found that the average of students completing the readings before class in eight psychology courses more than
doubled (from less than a third to almost 70%) if the material was to be included on an upcoming test. Finally, Ryan
(2006) and Rodríguez Prieto (2008) developed focused homework assignments based on the assigned reading,
which could also serve as study guides later in the semester. Students who had the focus worksheets with teacher
comments (Ryan, 2006) or who used them at the beginning of class to start the class discussions (Rodríguez
Prieto, 2008) performed better on the regularly scheduled exams than those who did not.
The current study aimed at increasing reading compliance in FL courses so that teachers can devote more
class time to quality practice and less time to redundant grammar explanations. Daily quizzing was incorporated
in two intermediate Spanish courses because, as described before, it has consistently been proved to rise reading
compliance among college students.
1.2. The use of clickers
As mentioned earlier, Millennial students have a strong interest in using technology in every aspect of their
lives. But technology is not only an asset for them but also a learning preference, whose other learning preferences
include structure as well as entertainment and excitement (Raines, 2002). Explorations in the field of FL teaching
using technology is a reasonable way of motivating and connecting with the new waves of students entering
college.
Beekes (2006: 33) recognized that the potential benefits of using clickers for class tests were still unknown and
Reese (2010) suggested that a useful tech idea could be integrating clickers to assess knowledge. The current
study aims to fill that void in the field of FL learning by using quantitative as well as qualitative data to evaluate
the impact this technology might have for assessing the contents of grammar and vocabulary classes via regular
quizzing.
Clickers are wireless personal response systems which typically resemble a TV remote (see Figure 1) and that
each student can use to select the answer for any question posed during a lecture. A small receiver connected to
the instructor’s PC collects all students’ votes which can be immediately displayed onscreen as a bar graph. Once
the instructor is familiar with the system, it literally requires a couple of minutes to plug the receiver to the PC and
open a new session. Some current models do not require the use of specific software to write the questions, and
that allows the instructors to pose questions in several ways: in a Powerpoint© slide, while watching a video on
YouTube©, and alike. This technology ensures anonymity to the students’ responses, allows for the comparison of
student’s votes with those from the rest of the class, and can be used to provide immediate feedback to the class
RLyLA Vol. 09 (2014), 88-96 | 89
Figure 1. Clicker used in the current study.
The pedagogical value of clickers in the academic setting has been studied in a diversity of disciplines whose
common denominator was a large number of students per section. The vast majority of these studies (Hake, 1998;
Burnstein and Lederman, 2001; Jones, et al., 2001; Elliott, 2003; Wit, 2003; Draper and Brown, 2004; Kennedy and
Cutts, 2005; Beekes, 2006, 2009; Bunce, et al., 2006; Ewing, 2006; Freeman, et al., 2006, 2007; Lee and Bainum,
2006; Narloch, et al., 2006; Poirier and Feldman, 2007; Stowell and Nelson, 2007; Cotner, et al., 2008; King and
Joshi, 2008; Morling, et al., 2008; Gier and Kreiner, 2009; Mayer, et al., 2009; Marlow, 2010) provide converging
evidence that students as well as professors agree that clickers afford the following benefits: a) they increase
motivation and interest for the class, b) they increase participation and involvement, c) they allow students to
self-assess and compare their performance to the rest of the class, d) they provide immediate feedback, e) they
are easy and fun to use, f) they can provide anonymity of the students’ response, and, more importantly, g) they
are believed to contribute to learning. This short outline of the many benefits for using clickers should be enough
to justify their embracing; however, a recurrent limitation from many of these studies is that they simply rely
on qualitative methods such as questionnaires, interviews, and classroom observations. Accordingly, there is a
decisive need to conduct more studies using quantitative methods to validate these encouraging results.
In the field of SLA, the use of clickers has been barely investigated. There are only four published studies (Cutrim
Schmid, 2007, 2008; Cardoso, 2010, 2011), all of which relied exclusively in qualitative methods and did not test the
effectiveness of clickers in producing learning gains. Cutrim Schmid (2007) used clickers with 30 English learners
and collected data from classroom observations, questionnaires, and oral interviews with the students. She found
out that the technology provided immediate feedback the participants could use for self-assessment and selfesteem, as they could compare their progress in relation to the rest of the class. Cutrim Schmid (2008) used the
same data as in her previous study but focused on interactivity or increased interaction and she discovered that
clickers increased the scope of interactivity as students participated more actively. But a closer look at the videos
also showed that the level of interactivity was relatively shallow as students provided answers to the questions
posed but rarely justified their answers, forcing the teacher to comment on their responses. Similarly, Cardoso
(2010, 2011) examined the perceptions of 30 English learners on the use of clickers via questionnaires and oral
interviews. Participants agreed for the most part that clickers were a positive addition to the class, increased their
participation and enjoyment of the classes, allowed them to self-assess and compare their performance with that
of their peers, fostered interaction, and contributed to learning. But none of these four studies really quantified
whether the use of clickers had an effect on FL learning outcomes such as better grades.
The rationale for choosing clickers for the daily quizzes in the present study was twofold: a) to allow Millennials
to receive immediate feedback for their responses, as one of their main liabilities is zero tolerance for delays, and b)
to allow them to compare their responses to those of their classmates, information they could use to increase their
self-esteem. But the main goal was to check the effectiveness of using clickers in increasing the likelihood that
students read the material and have some prior knowledge of the day’s topic, given that today’s students spend
many hours a day interacting with technology, which is also one of their learning preferences.
The present study addressed the following research questions:
1. What is the effect of using clickers on students’ compliance with the reading assignments, as measured
by regular quiz scores and their reported preparedness?
2. What are the students’ opinions about the use of clickers to increase their preparedness, based on their
preferred delivery method for quizzes and their comments on the use of this technology?
RLyLA Vol. 09 (2014), 88-96 | 90
The current study is breaking new ground in the field of SLA for being the first quantitative research measuring
the effect of clickers on learning gains. It is also one of the very few studies dealing with reading compliance in FL
courses, an often neglected area of research in SLA.
2. Method
2.1. Participants and setting
The study was conducted in a state-run public university serving more than 22,000 students in Muncie, Indiana.
38 college-level students of L2 Spanish enrolled in the second semester of second-year Spanish participated in
the study. All participants were recruited from two intact Spanish classes. Each section was randomly assigned a
treatment condition at the start of the semester: the Clicker group, who used clickers to buzz in their responses
from the daily quizzes (n = 19), and the Paper and Pencil group, who used the traditional method of writing their
responses in quizzes (n = 19).
Participants met for 75 minutes twice a week, with the same instructor, who is also the author of this study,
covering the same content by using the exact same materials and syllabus. Participants’ ages ranged from 18 to
22 with a mean age of 19.2 years. There was a similar distribution by gender, with 18 males and 20 females. Most
of the participants were freshmen, 22 of them, with 9 sophomores, 5 juniors, and 2 seniors. More than three fourths
(76.3%) were taking the class to fulfill the university requirement but the great majority of them (81.6%) also had
plans to continue studying Spanish after that class.
2.2. Instruments and data collection
All students were required to purchase a clicker for the whole semester, irrespective of the treatment condition.
Clickers were used in one or two activities per class session to review the contents of the class and to check the
students’ progress, as well as to promote discussions in the FL.
During the semester, all participants were reminded about the following’s class reading assignment with the
textbook pages written on the board. And at the beginning of each of the 16 classes with an assigned vocabulary
or grammar reading section, all students had a quiz about the contents of the reading, for a 10% of the final grade
and as stated in the syllabus. Each quiz had 10 multiple-choice questions, e.g., 5 or less options per question, true/
false, matching… Most questions were fill-in-the-gap Spanish sentences or some information about the reading in
Spanish. Here is an example: Dudo que mis hijos __________ mucha televisión durante la semana. a) mirando, b)
miran, c) miren, d) mires. Quizzes generally lasted no more than 10 minutes.
Participants in the Clicker group took the daily quizzes using their clickers. They had no time restrictions to
vote for each question and once all students voted, which could be checked in a small display box onscreen, their
voting results were displayed in a bar graph, hence receiving immediate feedback about their performance. There
was time for some discussion about their results after completing all the questions. Participants in the Paper and
Pencil group completed the exact same quizzes but in a piece of paper and they received the corrections and
grades in the next class.
During the last 25 minutes of a class in the last week of the semester, the instructor left the classroom and
students received a questionnaire with a variety of questions about some personal information, their reported
preparedness for class, their opinions about clickers, and their preferred quiz delivery method.
2.3. Data analysis
All statistical analyses were carried out with SPSS, v.17.0. for Windows. The alpha level for significance was
set at p < 0.05. An independent-samples t test was used to compare the participants’ average quiz score taking
the use of clickers (or not) for the quizzes as the grouping variable. Frequencies and descriptive statistics were
used elsewhere.
3. Results
The first research question examined whether clickers had any effect on reading compliance or students’
preparation for each class. In order to answer that question, participants’ reported preparedness for class during
the semester was collected and analyzed as well as their averaged grade on the 16 daily quizzes.
Participants’ averaged grade on the 16 quizzes was calculated. In accordance with previous research (ConnorGreene, 2000; Thorne, 2000; Ruscio, 2001; Clump, et al., 2004), reading compliance was around two thirds (n = 38,
RLyLA Vol. 09 (2014), 88-96 | 91
M = 65.132, SD = 13.920), given that they were regularly quizzed during the semester. However, the students
in the Paper and Pencil condition obtained greater scores on the averaged quiz grades (n = 19, M = 67.368,
SD = 16.575) than those in the Clicker condition (n = 19, M = 62.895, SD = 10.631). The quiz grades were analyzed
by comparing the groups using an independent-samples t test. There was a significant result (n = 38, t = -.990, df
= 36, p < 0.05) between the groups, indicating that students in the Clicker condition obtained significantly lower
quiz grades during the semester than those in the Paper and Pencil condition. These data suggest that students
who completed the quizzes in a piece of paper obtained better scores on the regular quizzes than those who took
the quiz using the clickers.
Students were asked in the questionnaire if they prepared on a regular basis before taking the quizzes. Results
indicated that about two thirds of the students in each class recognize preparing for the quizzes (n = 38, M = 65.8,
SD = 0.481). But it was those participants in the Clicker condition, as shown in Figure 2, who prepared for the
quizzes more often, despite obtaining significantly lower scores on the quizzes.
100
80
73.7% (14/19)
57.9% (11/19)
60
40
20
0
Clicker
Paper and Pencil
Figure 2. Reported preparation for the quizzes on a daily basis (n = 38).
The second research question investigated whether participants preferred the clickers for taking the quizzes
or the traditional paper and pencil format. Their opinions about the use of clickers were gathered with the help of
statements they could agree or disagree with, and by open-ended questions in the questionnaire.
Students indicated in the questionnaire whether their preferred method for taking quizzes was in a piece of
paper or with clickers, irrespective of how quizzes were delivered to them during the semester. As shown in Figure
3, results indicated that clickers were the main preferred delivery method only for those participants who actually
used them for that purpose during the semester, i.e., only for students in the Clicker group (18/19, 94.7%), but not
for those in the Paper and Pencil group (5/19, 26.3%).
100
94.7% (18/19)
80
60
40
26.3% (5/19)
20
0
Clicker
Paper and Pencil
Figure 3. Clickers as the preferred delivery method for the daily quizzes (n = 38).
When participants were asked to explain why they preferred such quiz delivery method, most participants who
chose clickers reported that clickers were easy to use while at the same time they got their results right away,
receiving immediate feedback on their answers. Here are two verbatim samples from the dataset:
•
It’s easier and get’s you results faster. Also allows instructor to give reasons for the answer.
•
It is much simpler and graded right there and get explanation as you go, not 2 days from the quiz, also
very interactive.
RLyLA Vol. 09 (2014), 88-96 | 92
On the other hand, students who preferred to complete the quizzes on a piece of paper mentioned that they
chose that method because it allowed them to take their time and to focus on each individual question. They
also mentioned that this method provided them with the ability to refer to previous answers and make changes if
necessary. Two verbatim samples from the dataset include:
•
You have more time and no pressure, and you can go back & change answers.
•
When the quiz is on a paper I can easily review, or change my answers, as well as skip and return to
answers that are hard.
Students who used the clickers to complete the quizzes were asked to agree or disagree with six statements
about that tool. In addition to those statements, participants in the Clicker Condition were asked to comment on
the usefulness of the tool for taking the quizzes. Results indicated that participants mostly agreed with all the
proposed statements about clicker use but for the fifth statement, which established that they studied more often
because they took the quizzes with the clicker (M = 52.8, SD = 0.506). More specifically, they felt clickers made
quizzes easier to complete (M = 97.2, SD = 0.167), were useful for that purpose (M = 97.2, SD = 0.167), enjoyed
using them for the quizzes (M = 88.9, SD = 0.319), granted them reasonable time to respond to each question
(M = 83.3, SD = 0.378), and would like to use them again in future Spanish courses (M = 86.1, SD = 0.351). This
indicates that no matter how much they enjoyed using clickers for the quizzes, or how useful and easy they were
to complete the quizzes while allowing them enough time to answer to the questions, still they will not study more
often nor prepare a little bit more on a regular basis to obtain higher scores in future quizzes.
When participants were asked to provide some comments on the usefulness of taking quizzes using clickers,
most participants agreed that they were helpful, easy and simple to use, provided immediate feedback about
the right answer, and were a faster method than paper and pencil. The very few negative comments received
mentioned that sometimes they felt rushed to provide an answer (even though there was no time limit imposed to
buzz in an answer) and that clickers were easy to lose or forget at home.
4. Discussion
The present data show that students in the intermediate Spanish classes under study prepared for class
regularly, as measured by their average grade on the daily quizzes, which ranged from 62.9 in the Clicker group
to 67.4 in the Paper and Pencil group. Those averages are more than double of the students who regularly come
prepared to class when they do not have quizzes, which typically represent about a third of the class (Burchfield
and Sappington, 2000; Sikorski, et al., 2002; Clump, et al., 2004). Without a control group covering the same
contents but not taking the quizzes, it cannot be ensured that those positive averages were mostly the direct
result of the quizzes. However, there is enough converging evidence about the positive effect of regular quizzing
on students’ increased preparation in other disciplines (Connor-Greene, 2000; Thorne, 2000; Ruscio, 2001; Clump,
et al., 2004) as not to expect the current results or to deny the positive effect of quizzes on reading compliance.
This is the first quantitative study measuring the effect of clickers on students’ average quiz grades during a
semester and in the field of SLA. The independent-samples t test results indicated that clickers affected negatively
these scores, as students in the Clicker group obtained significantly lower scores on the quizzes than the Paper
and Pencil group. One of the reasons for this result is that this technology by itself did not encourage participants
to study more often prior to the quizzes, as evidenced by the participants’ low agreement level (52.8%) on the
statement “I studied more frequently because I took the quizzes with the clicker.” Participants’ lower scores on the
quizzes may be explained in part by the limitations imposed by this technology, such as the inability to go back
and correct answers in more difficult questions. Although there was not a time limit to select the correct answer for
each question, some students may have felt pressured to respond when they saw most of their peers had already
voted. The immediate feedback and the possibility of comparison with the rest of the class, which at first were the
main benefits of this technology to increase students’ self-esteem and willingness to comply with the assigned
readings, may actually provoked the opposite effect by pressuring and frustrating students who needed to think
their responses more carefully or who would have preferred to go back and forth before making a final decision in
the hardest questions. This technology mainly helped in the collection and grading of students’ responses, taking
into account Millennials lack of tolerance for delays, their need of immediate feedback, and their passion for all
things technology. Unfortunately, our quantitative results did not extend to class tests the converging qualitative
positive benefits on adopting clickers in FL classes found in other areas such as self-assessment, self-esteem,
interactivity, participation, and enjoyment (Cutrim Schmid, 2007, 2008; Cardoso, 2010, 2011).
Besides the negative effects on students’ quiz scores, impressionistic students’ behavior throughout the
semester in relation to the potential increased participation, voluntariness, and class involvement thanks to the
use of clickers was anything but positive. Aside from quizzing, participants in both groups seemed to associate
RLyLA Vol. 09 (2014), 88-96 | 93
class participation with buzzing in an option for the daily clicker activities and then remaining passive. More
silences were observed in response to a question if it was displayed onscreen and students used clickers for their
answers than if they did not use clickers. This is exactly what previous research also found, that reluctant and
quiet students did not behave differently after using the clickers (Jones, et al., 2001) or that shy students feel the
technology did not allow them to express their opinions (Lee and Bainum, 2006). In sum, the more this technology
was incorporated in the classroom dynamic for regular activities and quizzing, the more evident it became that it
was not producing the expected effects.
A final comment should be made about the observation that our students mostly preferred the quiz delivery
method to which their group was assigned. This indicates that students in the Clicker group were not aware of
the negative effect clickers had on their quiz performance. Lantz (2010) previously stated that students generally
do not realize how effective any given learning strategy might be. Specifically, he mentioned that clickers may
affect student learning without the students realizing their effectiveness, to which I add that this is even possible
if the effect is negative, which is more worrying. On the other hand, students may simply be satisficing with the
researcher, a possible limitation of the present study to which we turn in the following section.
5. Limitations and Future Research
This study was limited in several ways. As in all research on teaching techniques, variables other than those
included in the present study may have influenced the results. The number of students in each section was low
and there was not a control group without regularly-scheduled quizzes to compare their reading compliance with
that from the experimental groups.
Having the main author as the instructor for the experimental groups was a potential source of subjectivity,
but active measures were taken to minimize this effect: a) by using the same teaching and testing materials,
b) by covering the same contents per class session, c) by leaving the class while participants completed the
questionnaires, d) by assigning participants to a randomly-selected experimental condition based on two intact
classes at the same level during the same semester and at the same institution, and similar. On the other hand, this
also ensured the uniformity of the teaching and testing materials and the consistency on the amount and quality
of content covered during the semester, which would have been difficult to ensure with different instructors even
if sharing the same syllabus and textbook.
When measuring subjective and attitudinal variables, participants select responses that are socially acceptable
and according to how they would like to portray themselves, with a tendency for self-enhancement (Oller, 1981,
1982; Krueger, 1998; Sappington, et al., 2002). Nonetheless, the propensity for self-enhancement and satisficing
with the researcher (Krosnick, 1999) was minimized by taking the following actions: a) the instructor left the
classroom during data collection and a peer delivered and collected the questionnaires; b) the volunteer instructor
read an introductory paragraph to the participants stating that their responses will be sealed in an envelope and
will not be delivered to the instructor until final grades were submitted to the university; and c) the instructions in
the questionnaire asked participants to be sincere and give the response that best described them, not what they
thought should be the best answer.
Despite these limitations, it should be recognized that these findings are strong in that the effect of clickers
on FL reading compliance was attested via different statements on the questionnaire, the participants’ opinions
and comments, and, more importantly, their averaged quiz grades during the semester. This is the first exploratory
and quantitative study in SLA measuring the effectiveness of clickers to assess knowledge and boost reading
compliance. Ideas for future quantitative research studies include the impact of clickers in the classroom dynamics,
such as willingness to communicate in the FL after voting or the amount of sentences students provide in the FL
if they were asked to buzz in a single option after a discussion in small groups.
6. Conclusions
This initial exploration on the effectiveness of clickers on students reading compliance indicated that the use of
clickers affected negatively students’ performance on quizzes as the tool did not facilitate access to the readings
neither reminded them about the reading assignments nor motivated them to try and best their scores for the
following classes, despite providing them with immediate feedback and the ability to compare their scores with
those of their classmates. More quantitative research studies are needed until these results can be generalized to
the population of FL learners.
RLyLA Vol. 09 (2014), 88-96 | 94
REFERENCES:
Beekes, W. (2006). “The ‘millionaire’ method for encouraging participation”, Active Learning in Higher Education,
7/1: 25-36. doi:10.1177/1469787406061143
Beekes, W. (2009). ““Is that your final answer?” Encouraging student participation using the personal response
system”, in Enhancing Student-Centred Learning in Business and Management, Hospitality, Leisure, Sport,
and Tourism. Oxford: The Higher Education Academy, Oxford Brookes University. pp. 76-86.
Bunce, D., VandenPlas, J., and Havanki, K. (2006). “Comparing the effectiveness on student achievement of a
student response system versus online WebCT quizzes”, Journal of Chemical Education, 83/3: 488-493.
doi:10.1021/ed083p488
Burchfield, C., and Sappington, J. (2000). “Compliance with required reading assignments”, Teaching of Psychology,
27/1: 58-60.
Burnstein, R., and Lederman, L. (2001). “Using wireless keypads in lecture classes”, The Physics Teacher, 39/1:
8-11. doi:10.1119/1.1343420
Burroughs, N., Kearney, P., and Plax, T. (1989). “Compliance-resistance in the college classroom”, Communication
Education, 38/3: 214-229. doi:10.1080/03634528909378758
Cardoso, W. (2010). “Clickers in foreign language teaching: A case study”, Contact, 36/2: 36-55.
Cardoso, W. (2011). “Learning a foreign language with a learner response system: The students’ perspective”,
Computer Assisted Language Learning, 24/5: 393-417. doi:10.1080/09588221.2011.567354
Clump, M., Bauer, H., and Bradley, C. (2004). “The extent to which psychology students read textbooks: A multiple
class analysis of reading across the psychology curriculum”, Journal of Instructional Psychology, 31/3:
227-232.
Connor-Greene, P. (2000). “Assessing and promoting student learning: Blurring the line between teaching and
testing”, Teaching of Psychology, 27/2: 84-88. doi:10.1207/S15328023TOP2702_01
Cotner, S., Fall, B., Wick, S., Walker, J., and Baepler, P. (2008). “Rapid feedback assessment methods: Can we
improve engagement and preparation for exams in large-enrollment courses?”, Journal of Science Education
and Technology, 17/5: 437-443. doi:10.1007/s10956-008-9112-8
Cutrim Schmid, E. (2007). “Enhancing performance knowledge and self-esteem in classroom language learning:
The potential of the ACTIVote component of interactive whiteboard technology”, System: An International
Journal of Educational Technology and Applied Linguistics, 35/2: 119-133. doi:10.1016/j.system.2007.01.001
Cutrim Schmid, E. (2008). “Using a voting system in conjunction with interactive whiteboard technology to enhance
learning in the English language classroom”, Computers & Education, 50/1: 338-356. doi:10.1016/j.
compedu.2006.07.001
Draper, S.W., and Brown, M.I. (2004). “Increasing interactivity in lectures using an electronic voting system”,
Journal of Computer Assisted Learning, 20/2: 81-94. doi:10.1111/j.1365-2729.2004.00074.x
Elliott, C. (2003). “Using a personal response system in economics teaching”, International Review of Economics
Education, 1/1: 80-86.
Ewing, A.T. (2006). “Increasing classroom engagement through the use of technology”, in http://mcli.maricopa.
edu/files/mil/ewing_rpt.pdf [accessed: 5.24.2012].
Freeman, M., Blayney, P., and Ginns, P. (2006). “Anonymity and in class learning: The case for electronic response
systems”, Australasian Journal of Educational Technology 22/4: 568-580.
Freeman, S., O’Connor, E., Parks, J., Cunningham, M., Hurley, D., Haak, D., Dirks, C., and Wenderoth, M. (2007).
“Prescribed active learning increases performance in introductory biology”, CBE Life Sciences Education,
6/2: 132-139. doi:10.1187/cbe.06-09-0194
Gass, S. (1997). Input, Interaction and the Second Language Learner. Mahwah, NJ: Lawrence Erlbaum Associates.
Gass, S., and Selinker, L. (2008). Second Language Acquisition: An Introductory Course (3rd ed.). Mahwah, NJ:
Lawrence Erlbaun Associates.
Gier, V., and Kreiner, D. (2009). “Incorporating active learning with Powerpoint-based lectures using content-based
questions”, Teaching of Psychology, 36/2: 134-139. doi:10.1080/00986280902739792
Hake, R. (1998). “Interactive-engagement versus traditional methods: A six-thousand-student survey of mechanics
test data for introductory physics courses”, American Journal of Physics, 66/1: 64-74. doi:10.1119/1.18809
Howe, N., and Strauss, W. (2000). Millennials Rising. New York: Vintage Books.
Jones, C., Connolly, M., Gear, A., and Read, M. (2001). “Group interactive learning with group process support
technology”, British Journal of Educational Technology, 32/5: 571-586. doi:10.1111/1467-8535.00226
Karp, D., and Yoels, W. (1976). “The college classroom: Some observations on the meanings of student
participation”, Sociology and Social Research, 60/4: 421-439.
RLyLA Vol. 09 (2014), 88-96 | 95
Kennedy, G., and Cutts, Q. (2005). “The association between students’ use of an electronic voting
system and their learning outcomes”, Journal of Computer Assisted Learning, 21/4: 260-268.
doi:10.1111/j.1365-2729.2005.00133.x
King, D., and Joshi, S. (2008). “Gender differences in the use and effectiveness of personal response devices”,
Journal of Science Education and Technology, 17/6: 544-552. doi:10.1007/s10956-008-9121-7
Krosnick, J. (1999). “Survey research”, Annual Review of Psychology, 50/1: 537-567. doi:10.1146/annurev.
psych.50.1.537
Krueger, J. (1998). “Enhancement bias in descriptions of self and others”, Personality and Social Psychology
Bulletin, 24/5: 505-516. doi:10.1177/0146167298245006
Lantz, M. (2010). “The use of ‘clickers’ in the classroom: Teaching innovation or merely an amusing novelty?”,
Computers in Human Behavior, 26/4: 556-561. doi:10.1016/j.chb.2010.02.014
Lee, J., and Bainum, C. (2006, April). “Do clickers depersonalize the classroom? An evaluation by shy students”,
poster session presented at the annual meeting of the Western Psychological Association, Palm Springs, CA.
Marlow, D. (2010). “Engaging syntax: Using a personal response system to encourage grammatical thought”,
American Speech, 85/2: 225-237. doi:10.1215/00031283-2010-012
Mayer, R., Stull, A., DeLeeuw, K., Almeroth, K., Bimber, B., Chun, D., Bulger, M., Campbell, J., Knight, A., and
Zhang, H. (2009). “Clickers in college classrooms: Fostering learning with questioning methods in large
lecture classes”, Contemporary Educational Psychology, 34/1: 51-57. doi:10.1016/j.cedpsych.2008.04.002
Morling, B., McAuliffe, M., Cohen, L., and DiLorenzo, T. (2008). “Efficacy of personal response systems
(“clickers”) in large, introductory psychology classes”, Teaching of Psychology, 35/1: 45-50.
doi:10.1080/00986280701818516
Narloch, R., Garbin, C., and Turnage, K. (2006). “Benefits of prelecture quizzes”, Teaching of Psychology, 33/2:
109-112. doi:10.1207/s15328023top3302_6
Oller, J.W., Jr. (1981). “Can affect be measured?”, International Review of Applied Linguistics, 19/3: 227-235.
Oller, J.W., Jr. (1982). “Gardner on affect: A reply to Gardner”, Language Learning, 32/1: 183-189.
doi:10.1111/j.1467-1770.1982.tb00526.x
Poirier, C., and Feldman, R. (2007). “Promoting active learning using individual response technology in large
introductory psychology classes”, Teaching of Psychology, 34/3: 194-196. doi:10.1080/00986280701498665
Prensky, M. (2012). From Digital Natives to Digital Wisdom: Hopeful Essays for 21st Century Learning. Thousand
Oaks, CA: Corwin.
Raines, C. (2002). “Managing millennials”, in http://www.generationsatwork.com/articles/millenials.htm [accessed:
5.29.2012].
Reese, S. (2010). “Teaching languages to the millennial generation”, The Language Educator, 5/4: 34-38.
Rodríguez Prieto, J.P. (2008). “Creación de preguntas de preparación para promover las lecturas asignadas en
cursos de contenido en lenguas extranjeras”, Revista Iberoamericana de Educación, 46/3: 1-10.
Ruscio, J. (2001). “Administering quizzes at random to increase students’ reading”, Teaching of Psychology, 28/3:
204-206. doi:10.1207/S15328023TOP2803_08
Ryan, T. (2006). “Motivating novice students to read their textbooks”, Journal of Instructional Psychology, 33/2:
135-140.
Sappington, J., Kinsey, K., and Munsayac, K. (2002). “Two studies of reading compliance among college students”,
Teaching of Psychology, 29/4: 272-274. doi:10.1207/S15328023TOP2904_02
Sikorski, J., Rich, K., Saville, B., Buskist, W., Drogan, O., and Davis, S. (2002). “Student use of introductory texts:
Comparative survey findings from two universities”, Teaching of Psychology, 29/4: 312-313. doi:10.1207/
S15328023TOP2904_13
Spodark, E. (2010). “Structuring a language course to respond to millennial generation workplace characteristics”,
The Language Educator, 5/4: 39-42.
Stowell, J., and Nelson, J. (2007). “Benefits of electronic audience response systems on student participation,
learning, and emotion”, Teaching of Psychology, 34/4: 253-258. doi:10.1080/00986280701700391
Swain, M. (1985). “Communicative competence: Some roles of comprehensible input and comprehensible output
in its development”, en S. Gass, and C. Madden (eds.) Input in Second Language Acquisition, Rowley, MA:
Newbury House. pp. 235-253.
Swain, M. (1995). “Three functions of output in second language learning”, en G. Cook, and B. Seidlhofer (eds.)
Principle and Practice in Applied Linguistics, Oxford, England: Oxford University Press. pp. 125-144.
Thorne, M. (2000). “Extra credit exercise: A painless pop quiz”, Teaching of Psychology, 27/3: 204-205. doi:10.1207/
S15328023TOP2703_09
Wit, E. (2003). “Who wants to be… The use of a personal response system in statistics teaching”, MSOR
Connections, 3/2: 14-20. doi:10.11120/msor.2003.03020014
RLyLA Vol. 09 (2014), 88-96 | 96
Vol. 9 año 2014, 97-105
EISSN 1886-6298
HOW AN IDEA GERMINATES INTO A PROJECT OR THE INTRANSITIVE RESULTATIVE
CONSTRUCTION WITH ENTITY-SPECIFIC CHANGE-OF-STATE VERBS1
Andreea Rosca
Centro Universitario de la Defensa (Zaragoza)
Abstract: This study discusses how seven of Levin’s (1993) entity-specific change-of-state verbs (i.e. bloom, blossom, flower,
germinate, sprout, swell, and blister) are subsumed into the intransitive resultative construction by highlighting and making use
of the external and internal constraints proposed by the Lexical Constructional Model (LCM; Ruiz de Mendoza and Mairal 2007).
External constraints refer to cognitive mechanisms, such as high-level metaphor and/or metonymy whereas internal constraints
are concerned with the encyclopedic and event structure makeup of verbs. The Internal Variable Conditioning constraint is at
work when the information encapsulated by a predicate determines the choice of the Z element in an intransitive resultative
construction. The semantic makeup of the verb swell and the entity undergoing swelling constrain the nature of the resultant
entity Z which must be bigger in size or have a bigger value than the Y element (e.g. The work, which was originally meant to
consist only of a few sheets, swelled into ten volumes).
Keywords: entity-specific change-of-state verbs, intransitive resultative construction, Lexical Constructional Model, external and
internal constraints, the Internal Variable Conditioning constraint.
1. introduction
The perspective adopted in this article inscribes itself in the field of Construction Grammar(s) (CxG), which
has been denominated by Östman and Fried (2004: 1) ‘family of Construction Grammars’ owing to its expanding
inventory of connected models. Within the broad framework of CxG, we have decided to base our present research
on the tenets formulated by the Lexical Constructional Model or LCM (Mairal and Ruiz de Mendoza 2008, 2009;
Ruiz de Mendoza and Mairal 2008, 2011; Ruiz de Mendoza 2013), for several reasons: (i) the LCM is a solid
theoretical model which strikes a balance between the roles of verbal semantics and constructions; (ii) the LCM,
in contrast to other lexical-constructional approaches (e.g. Iwata 2005; Nemoto 2005; Boas 2008), gives more
prominence to empirically validated cognitive notions, such as conflation, high-level metonymy and metaphor;
(iii) unlike other CxG approaches, the LCM distinguishes different levels of meaning representation; (iv) a major
advantage of this model is its intention to connect the linguistic realm with the computational one by joining
forces with the Artificial Intelligence project FunGramKB; and (v) the LCM studies the principles that regulate the
interaction between lexical items and constructions, which is precisely our main interest in this research.
In this article we examine the factors that license or block out the fusion between entity-specific changeof-state verbs and the intransitive resultative construction, by focusing only on seven members of this verbal
class (i.e. bloom, blossom, flower, germinate, sprout, swell, and blister) due to space limitations. The intransitive
resultative construction is a pattern that has been understudied by construction grammarians. We have also
selected this constructional pattern in order to show that this verb class exhibits a richer distributional range than
has been attested in the literature (Levin 1993, 2006; Wright 2002). Thus, the constructional pattern of these verbs
is not restricted to the inchoative/causative alternation.
1
Center for Research in the Applications of Language (www.cilap.es). Financial support for this research has been provided by the Spanish Ministry of Science and
Innovation, grant no. FFI2010-17610
| 97
Andreea Rosca
How an idea germinates into a project or the intransitive resultative construction with entity-specific change-of-state verbs
Before embarking upon the analysis of entity-specific change-of-state verbs, a clear distinction should be
made within the resultative construction between two change schemas. Ruiz de Mendoza and Luzondo (2011)
put forward the general principle of Resultatives under one common denominator which explains the chaotic
realization of end-results by means of two simple change schemas, i.e. A>A’ and A>B schemas. The first illustrates
that an entity A acquires a new property but retains its essence whereas the second indicates that an entity A
experiences a conspicuous change which leads to a loss of homogeneity or integrity. For example, a sentence
like Mary wiped the table clean falls into the first change schema (A>A’) because the patient (table) undergoes
a transformation of only one of its properties (e.g. from being dirty to being clean). The A>B change schema is
employed in the sentence The witch turned the boy into a frog, where the result-state into a frog indicates that the
patient (the boy) has suffered a total transformation, reaching a completely different state (i.e. from a human being
to an animate entity). In this article it will be made evident with corpora examples that these seven entity-specific
change-of-state verbs are subsumed into the A>A’ schema since they only describe the increase in size of a given
entity.
This article is organized as follows. Section 2 offers a brief overview of the theoretical framework of the LCM
and its main strengths. Section 3 introduces the reader to the intransitive resultative construction and stresses the
conceptual similarity between the verbs under scrutiny. Section 4 and 4.1 illustrate how different forms taken by
the intransitive resultative construction are motivated by the external and internal constraints postulated by the
LCM. The final section summarizes the main findings of this research study.
2. WHY THE LEXICAL CONSTRUCTIONAL MODEL?
The LCM stems from the Functional Lexematic Model (FLM; Martín Mingorance 1998), according to which
“lexical representations are the key as well as the source for predicting and explaining syntactic properties” (Faber
and Mairal 1999: 275). The FLM, which has been further developed by Faber and Mairal (1999), sets out to
investigate the paradigmatic structure of the lexicon by looking into semantic fields and classes and establishing
hierarchical structures on the basis of similarity and difference of meaning. In this connection, Faber and Mairal
(1999: 186) state that “verbs within the same subdomain have similar syntactic behavior”. This idea was also
put forward by Levin (1993: 5): “various aspects of the syntactic behavior of verbs are tied to their meaning.
Moreover, verbs that fall into classes according to shared behavior would be expected to show shared meaning
components”. Our findings related to entity-specific change-of-state verbs are in consonance with these authors’
claims, as will be seen in the following sections.
In addition, the LCM stands halfway between Role and Reference Grammar (Van Valin 2005) and constructionist
models of language (e.g. Goldberg 2006; Boas 2008) since, in contrast with cognitive theories, which ignore
the importance of verbs and place constructions above them, it claims that verbal semantics plays an active
role in determining meaning construction. It agrees with constructionism when stating that in a caused-motion
construction like They scorned him into depression the final meaning is provided by the construction itself and
cannot be derived from the predicate-argument structure of scorn. Nevertheless, the LCM cannot fully embrace
constructionist approaches since it is impossible for them to account for the broad array of constraints that are at
work in lexical-constructional fusion.
This model claims that the fusion processes between verbs and constructions are regulated by a set of internal
and external constraints. Internal constraints, which take into account the conceptual composition of lexical and
constructional configurations (i.e. their encyclopedic and event structure makeup), specify the conditions under
which a lexical predicate may modify its internal configuration so that it can become a candidate for subsumption
into a given construction. By contrast, external constraints, which usually take the form of high-level metaphor
and/or metonymy (see Ruiz de Mendoza and Mairal 2007), determine in what way or to what extent a lexical
predicate can be construed from a different perspective that may allow its meaningful integration into a given
construction without altering its internal structure.
Among the main reasons for the selection of the LCM as a theoretical framework for this study, we can mention
(see also Butler 2009: 26 for further discussion on the strengths of this model):
(i) Unlike Goldberg or Boas, who devote themselves exclusively to the examination of lower-level or highlevel schemas, the LCM embraces two other criteria for the taxonomy of constructions: idiomaticity/
eventivity (e.g. the resultative is an eventive construction whereas What’s X Doing Y? is an idiomatic
construction with fixed and variable elements) and meaning construction stratification (e.g. four levels of
meaning description: argument structure, implicational, illocutionary, discursive). This organization into
different levels of meaning description allows the LCM to study the way in which constructions from lower
levels are subsumed or integrated into higher-level constructions.
RLyLA Vol. 09 (2014), 97-105 | 98
Andreea Rosca
(ii) The LCM agrees with Boas on the importance of verbal semantics in meaning construction but acknowledges
the roles of metaphor and metonymy as constraining factors licensing or blocking subsumption.
(iii) Lastly, the LCM is currently being exploited in terms of computer-based implementations. The tenets of
this model are compatible with FunGramKB, which is an artificial intelligence knowledge base.
3. THE INTRANSITIVE RESULTATIVE CONSTRUCTION
The intransitive resultative construction is a fairly frequent type of configuration in our corpus with entity-specific
change-of-state verbs. The difference between the resultative (X CAUSES Y TO BECOME Z) and the intransitive
resultative (Y BECOMES Z) is marked by the presence of the X causal element in the first type of construction. In
the intransitive resultative construction the result seems to be obtained by the undergoer itself. The entity-specific
change-of-state verbs utilize either of two syntactic forms to convey an intransitive result: adjectival phrases (e.g.
[…] love-in-a-mist and forget-me-nots bloomed blue […]; Sketch engine doc#671530) or prepositional phrases
(e.g. Competition can deteriorate into rivalry; Sketch engine doc#79524).
A very peculiar way of codifying an intransitive result is exemplified by the sentence Her cheeks bloomed with
scarlet (Sketch engine doc#123606). The preposition with, which more readily expresses a cause (e.g. He died
with pneumonia), is used here to encode a result (cf. Her cheeks grew in beauty and as a result, they became
scarlet), which is licensed by the conceptual conflation of effects and causes, which underlies the activity of
the EFFECT FOR CAUSE metonymy. Folk knowledge based on misinterpreted perception may result in mixing
up effects and causes. For example, we can observe a dead body covered with skin lesions and erroneously
believe that the skin lesions have killed the person, which in fact are just a symptom of an underlying disease (a
bacterial infection). In a similar vein, we consider that the intransitive resultative construction is but a constructional
calque of the intransitive construction. The intransitive resultative construction is made possible by the high-level
metonymy A CHANGE OF STATE FOR A CAUSED EVENT. Consider the sentence The crops withered brown. This
linguistic expression designates a change of state, i.e. the crops becoming brown, but through world knowledge
we understand that this change of state happens by the action of what withers plants, namely certain weather
conditions. That is why we have a latent caused event.
The seven verbs that we are dealing with in this paper (i.e. bloom, blossom, flower, germinate, sprout, swell,
and blister) describe the coming into existence of an entity out of a pre-existent one. In this way, the blossoming
process of a flower refers to the development of a protuberance (bud/blossom) outside the stem of a plant (the
plant shifts from the vegetative to the reproductive stage). Although this process is generally viewed as a positive
change there might be some exceptions as can be seen in the sentences […] corruption bloomed in the worst
possible way (Sketch engine doc#1796738), Cysts germinate in the gastrointestinal tract […] (Sketch engine
doc#254041), or […] the tumor blossomed in a small cavity above the sinus […] (Sketch engine doc#758951). Also,
the verb blister, which refers to the causation of a swelling of the skin containing a watery fluid, can be regarded
as a negative change of state but this change does not threaten the “essence” of the experiencer. The verbs
bloom, blossom, flower can be exploited in a figurative way to refer to someone’s healthy, happy or successful
appearance probably because we associate a person’s glowing physical aspect with the positive emotions that
the sight and color of a blossoming flower transmit to us. It is also common knowledge that the flowering process
constitutes the maximum development of a plant and this stage can be reached only if the plant stays healthy. The
sentence The child blossomed into a good looking young man […] (Sketch engine doc#638230) is grounded in the
low-level metaphor REACHING ONE’S PRIME IS FLOWERING, whereby physical development of human beings is
conceptualized in terms of a plant reaching the blooming stage. This metaphor is subsidiary to a more generic one,
i.e. HUMANS ARE PLANTS, which in its turn is but a natural extension of the Great Chain of Being metaphor (cf.
Lakoff and Turner 1989), which attempts to comprehend human attributes and behavior through characteristics of
animals, plants, natural objects and artifacts. The life cycle or (physical/professional) development is regarded as
motion forth, e.g. people go from youth to old age, from a state of poverty to one of welfare just like a bud spreads
out of the plant to the surface in the sunlight (cf. the old Lakoffian metaphor PROGRESS IS MOTION). With respect
to germinate and sprout, these two verbs are similar since they make reference to the initial state of growth of a
seed, thus suggesting the beginning of progress. In the figurative domain, the appearance of shoots/buds/leaves
on a plant is correlated with the development of an idea/project/belief or the construction of buildings in a place
(e.g. Skyscrapers are sprouting up all over Europe). There is also an interesting implication about these verbs: since
early shoots or buds are usually a sign that there are prospects for a full-blown plant to emerge at some point
in time, they are a sign of hope (i.e. the prospects of a future fully fruit-bearing mature plant map onto the future
prospects for maturity of ideas, plans, etc., which are now at their initial stages). Both blister and swell describe
an increase in size or volume either of a body part (e.g. My feet and legs swell […]; Sketch engine doc#8227) or of
other kinds of surface (a blister also refers to a raised bubble on a painted or laminated surface).
RLyLA Vol. 09 (2014), 97-105 | 99
Andreea Rosca
4. SUBSUMPTION PROCESSES WITH THE ADJECTIVAL INTRANSITIVE RESULTATIVE CONSTRUCTION
As has been previously indicated, the intransitive resultative construction is a fairly productive configuration
with entity-specific change-of-state and especially the verbs under consideration. This construction can be
lexicalized by an adjectival phrase (AP), a prepositional phrase (PP) headed by to or into, a combination either of an
adverb and an adjectival phrase (Adv +AP) or of an adverb and a prepositional phrase (Adv+PP). The combinatorial
possibilities of these verbs are illustrated in the table below:
Table 1. The intransitive resultative construction with the verbs under scrutiny.
AP INTR. RES. CONSTRUCTION WITH THE VERBS UNDER SCRUTINY
But worst of all, the yellow flowers in the kitchen vase bloomed blue
Sketch engine
doc#88417
Luffa and sola plants still flowered a saffron yellow
Sketch engine
doc#1011853
[…] a series of concentric burns blistered black on a surface that stays white even in summer
Sketch engine
doc#2368880
PP INTR. RES. CONSTRUCTION WITH THE VERBS UNDER SCRUTINY
In his youth, the qualities of foresight and planning bloomed to perfection […]
Sketch engine
doc#101179
Instantly attracted to each other, this encounter blossomed into the most intense relationship of
Goldman’s life
Sketch engine
doc#255172
This idea of Canadian nationality later germinated into the 1947 Citizenship Act
Sketch engine
doc#969117
And as the spring came closer and closer, the tip nearest the ground swelled into a grotesque
head […]
BNC ACV 1184
Joseph of Arimathea is reputed to have planted one at Wearyall Hill, which subsequently
sprouted into the Glastonbury Thorn
BNC BMT 59
[…] the Grace Apartments on the city’s eastern edge blistered into a crime “hot spot” […]
COCA 2007
ADV+AP INTR. RES. CONSTRUCTION WITH THE VERBS UNDER SCRUTINY
Their throats [of roosters] would swell out big and then would come forth their booming challenge Sketch engine
[…]
doc#668491
ADV+PP INTR. RES. CONSTRUCTION WITH THE VERBS UNDER SCRUTINY
When, however, under her husband’s wing she had blossomed out into a lovely womanhood […]
Sketch engine
doc#645600
The adjectival phrase denotes an A>A’ type of change since it typically involves a change of a single property
of an entity, i.e. its color or size. Sometimes the adjectival phrase can be syntactically separated from the verb
by means of a preposition, either in or into. For the sake of clarity, consider the following examples listed below:
(1) a.[…] after winter rains the arid land bloomed in large patches of yellow, white and blue with the many
small flowers of wild adenostema, sage brush, ‘Spanish’ violets, shooting stars, mimulas and white
popcorn (Sketch engine doc#499614)
b. In the window sill the flowers of bygone days bloomed in motley green (Sketch engine doc#665157)
c. […] the Judas tree which grows in astonishing profusion, blossoming each spring into a vivid pink […]
(Sketch engine doc#385859)
d.When the sun broke through the clouds, the brown rock blossomed into earthy colors--ochre, siena,
umber, olive (Sketch engine doc#638039)
e.[…] brilliant trees flowered in a blaze of pure scarlet, and some in pure lavender […] (Sketch engine
doc#2273575)
f. In the freer spaces forget-me-nots flowered in nebulae, and dog-violets gave an undertone of dark
purple, with primroses for planets in the night (Sketch engine doc#2327813)
RLyLA Vol. 09 (2014), 97-105 | 100
Andreea Rosca
From a close inspection of examples (1)(a)-(f) we conclude that linguistic distance between the verb and
its adjectival specification of result, which is within the scope of the prepositional phrase, complies with the
part/whole affectedness principle and other dependency phenomena. At first, we tried to explain the difference
between a simple adjectival phrase resultative and the examples in (1) by looking at the indirect/direct causation
dichotomy. Thus, Fodor (1970) distinguishes between lexical causatives (e.g. Peter killed John) and analytical
causatives (e.g. Peter caused John to die). In his view, the former represent ‘atomic’ causal events whilst the latter
depict ‘compound’ causal events. He accounts for the analytical cause to construction in terms of a temporal
separation between the cause event and the effect event: “one can cause an event by doing something at a time
which is distinct from the time of the event” (ibid: 433). Lakoff (1987: 55) also argues that “the more direct the
causation, the closer the morphemes expressing the cause and the result” (cf. also Lakoff and Johnson 1980: Ch.
20; Haiman 1980). However, the indirect/direct causation distinction seems a rather implausible hypothesis for our
case. First of all, one cannot say that in (1a), for example, there is a greater time lapse between the blossoming
process and the coloring process just because the flowers are multi-colored: yellow, white, blue. Second of all, the
visual perception of the color occurs simultaneously with the blooming of the flower.
The difference between a sentence like The flower bloomed red and The flower bloomed in motley red lies in
the fact that the first one receives a whole-affectedness interpretation whereas the second one can be given a
part-affectedness reading. In the first example it is suggested that the surface of the flower becomes completely
red. The second sentence does not imply that only a small surface of the flower became red but that the color that
covers completely the surface of the flower is not homogenously distributed. The petals of the flower have elements
of great variety, thus, each hue occupies only a part of the surface of the flower. Also, the adjectival intransitive
resultative The flower bloomed red, which calls for a whole affectedness reading, is in clear contradiction with the
partial affectedness interpretation postulated by Broccias (2004: 109): “if an adjective in a resultative construction
describes a property P of an affected object Y, then P describes any part of Y (if possible)”. This generalization
would provide a convenient explanation for the ungrammaticality of a sentence like *He hammered the metal long/
tubular/square, where the adjectives long, tubular, and square cannot match with the resultative construction
simply because they describe properties of the whole entity. Luzondo (2011: 171) correctly points out that the
oddity of paraphrases like *We have drunk the barrels dry, but parts of it are wet, *John pushed the door open, but
part of it did not open throws doubt on the validity of Broccias’s part-whole affectedness generalization for the
resultative construction. Equally, the unacceptability of our own paraphrase *The flower bloomed red, but some
parts of it were yellow indicates that the color in the intransitive resultative construction is spread all over the
surface of the flower. Example (1a) clearly illustrates that the colors yellow, white, blue refer only to a small portion
of the surface of the arid land. In a similar vein, the colors in (1d), i.e. ochre, siena, umber, olive cover parts of
the surface of the rock. In (1f) the plural NP nebulae makes reference to a diffuse mass of interstellar dust or gas
which visually blends luminous patches with areas of darkness and hints again at the heterogeneity of the color
perception. In (1c) the color term pink becomes an NP by being incorporated into a relative clause, which probably
motivates the absence of a canonical intransitive resultative construction.
4.1. Subsumption processes with the prepositional intransitive resultative construction
The intransitive resultative construction with a prepositional phrase calls for an explanation based on what Ruiz
de Mendoza (2008) has labeled metaphoric amalgams. A metaphoric amalgam is a type of metaphoric interaction
which requires the integration of selected aspects from two or more metaphors that combine. There are two possible
ways in which metaphorical structure can combine, namely single-source metaphoric amalgams and double-source
metaphoric amalgams (cf. Ruiz de Mendoza 2008; Ruiz de Mendoza and Mairal 2011). Let us take into consideration
the following sentence: The concept bloomed into a debut cassette release […] (Sketch engine doc#446648). This
sentence is based on a double-source metaphoric amalgam, as can be observed in Table 2 below:
Table 2. The double-source metaphoric amalgam.
Source 
(natural process of blooming)
Target 
(change of state)
Source 
(change of location)
Flower
Concept
Source
Bloom
Process (development)
Motion
Blossom
Result (cassette)
Destination
A double source metaphoric amalgam involves two metaphoric sources that are mapped simultaneously onto
the same target domain. In our example two metaphors interact: A CHANGE OF STATE (OF AN ABSTRACT
ENTITY) IS BLOOMING and A CHANGE OF STATE IS A CHANGE OF LOCATION. Both metaphoric systems blend
into a more complex one in which ‘a concept’ undergoes a process of development understood in terms of selfinstigated motion from a source to a destination. The destination of motion is seen to coincide with the resultant
state of the abstract entity (‘cassette’).
RLyLA Vol. 09 (2014), 97-105 | 101
Andreea Rosca
At this point we would like to draw attention to the major role fulfilled by the lexical predicate in determining
the nature of its constructional arguments. For instance, the verb swell in the intransitive resultative construction
(Y BECOMES Z) designates the means by which transformation is achieved, i.e. physical expansion of a surface
or rise of position on a scale. The verb swell in the intransitive resultative construction obeys the Internal Variable
Conditioning constraint since the choice of the Z element is greatly constrained by the information encapsulated
by the verb swell (i.e. an entity becomes bigger in size or the value of the entity goes up on scale) and also by the
Y element. Thus, Z must be bigger in size or have a bigger value than Y. For validation purposes, consider the
following sentences:
(2)a. […] small settlements such as San Francisco swelled into cities (Sketch engine doc#194954)
b. This was the signal for a general clamour, which beginning in a low murmur gradually swelled into a great
noise in which everybody spoke at once (Sketch engine doc#458499)
c. The work, which was originally meant to consist only of a few sheets, swelled into ten volumes (Sketch
engine doc#643101)
d. Let a gale arise and swell into a storm, let a sea run that might appal the stoutest heart that ever beat
Examples (2a) and (2c) evoke the expansion schema whereby the Y element (small settlements and the
work consisting of a few sheets) increases in physical size until it becomes Z (a city or a work made up of ten
volumes). By contrast, (2b) and (2d) activate the intensity scale whereby the Y element (the low murmur and the
gale respectively) increases in intensity until it turns into Z (a great noise or a storm). All four examples observe the
A>A’ change schema. In (2a) and (2c) the city and the ten volumes work incorporate in their physical composition
the small settlements and correspondingly, the few sheets. The prepositional into phrase is used to realize the A’
element of the schema since English does not code a resultative adjective that captures the conceptual structure
called upon by into cities/a great noise/ten volumes/a storm. In cases like these, English makes figurative use of
the caused-motion construction (compare The blacksmith hammered the metal flat/into the shape of a fish/*into
a flat shape).
The intransitive resultative construction with bloom and blossom is regulated by the same Internal Variable
Conditioning constraint. The meaning of these verbs is more generic as it involves either that an entity becomes
bigger in size or goes from a lower-level stage of development to a higher-level stage of evolution, which does not
necessarily imply that this is positive. Let us take a look at some examples:
(3) a. Instantly attracted to each other, this encounter blossomed into the most intense relationship of
Goldman’s life (Sketch engine doc#255172)
b. What started as an entry-level job blossomed into a lifetime career and association with the University
c. Their partnership blossomed into marriage and their artistic union created one of the greatest vocal
phenomenon of this century (Sketch engine doc#554595)
d. Powell’s class project blossomed into a full-blown grant proposal […] (Sketch engine doc#569026)
e. […] Cedar Hill has blossomed from a rural town of about 6,800 in 1980 to a cosmopolitan area with a
population of almost 40,000 (Sketch engine doc#97371)
f. As China’s open door initiatives blossomed from slogan to reality at an astonishing rate, the gap between
Chinese statistical categories [..] imposed growing costs (Sketch engine doc#346706)
Sentences (3)(a)-(d) can be skeletally represented by the schema Y BECOMES Z. In these cases the Y element,
together with the verb blossom constrain the choice of the Z element which must be conceptually related to Y and
must involve a higher-level of development than Y. In turn, (3e) and (3f) display the semantics Y TURNS FROM S
TO Z, where Y = Cedar Hill/China’s open door initiatives, S = rural town/slogan, Z = cosmopolitan area/reality (i.e.
S = initial state; Z = final state).
Although (3)(a)-(f) constitute positive changes of state, the intransitive resultative construction with the verb
bloom can also encode negative end results as can be seen in the examples reproduced in (4)(a)-(d):
RLyLA Vol. 09 (2014), 97-105 | 102
Andreea Rosca
(4)a. As these growing gaps inevitably sow seeds of resentment among those less fortunate which perhaps
bloom into terrible acts (Sketch engine doc#472205)
b. WHAT TODAY MIGHT be seen as an isolated problem for a limited number of companies promises to
bloom into big trouble for us all (Sketch engine doc#566316)
c. The seeds have since bloomed into thousands of resistance fighters and foreign terrorists (Sketch engine
doc#593788)
d. Epiphanies don’t come much grander than that, and Shulgin’s interest in psychoactive drugs bloomed
into an obsession (Sketch engine doc#1734479)
The examples above exploit another meaning extension of the verb bloom that focuses not on the youthful
and vigorous aspects of blooming, but on the sudden appearance of the flower. It is a matter of attribute selection,
which is typical of metaphorical extension (in a metaphoric mapping not everything is mapped but, on the basis of
the Correlation Principle, only the source structure that best matches the implicational structure of the target; Ruiz
de Mendoza and Santibáñez 2003; Ruiz de Mendoza 2011).
As mentioned earlier, the intransitive resultative construction can display a compound result expressed either
by a combination between an adverb and an adjectival phrase or between an adverb and a prepositional phrase.
Let us take each case in turn. The sentence The balloon swells out tight and full (Sketch enginedoc#1041811)
combines the adverb out with two adjectival phrases, i.e. tight and full. Our example might seem to contradict
Goldberg’s (1991b: 368) Unique Path constraint, which stipulates that “if an argument X refers to a physical object,
then more than one distinct path cannot be predicated of X within a single clause”. This constraint has two main
entailments: (1) X cannot move to two different locations at a given time t; and (2) the motion must describe a
path within a single landscape. Thus, resultatives are believed to be incompatible with directional phrases (cf.
*Sam kicked Bill black and blue out of the room). Nevertheless, Goldberg’s formulation of this constraint does not
really explain why the constraint happens. The constraint is grounded in the physical impossibility of an integrated
object following two different paths at the same time. Alternatively, she postulates the Unique Change of State
constraint, according to which two distinct changes of state cannot be simultaneously predicated of an entity in
a single clause.
The adverb out normally describes a path (e.g. He went out). However, out in our example indicates a result
(swelling along the horizontal axis), which strictly speaking involves a path that is internal to the object. But there
is no motion along an external path, which is what Goldberg’s Unique Path constraint captures. Out indicates
external orientation whereas the composite adjectival phrase parametrizes the property acquired by the inflated
balloon. The adverb out evokes the surface expansion schema whilst the adjectival phrase refers to the size of an
entity. Also, the adjectival phrase tinges the intransitive resultative construction with telicity: the balloon inflates
until it becomes tight and full. The adjectival phrase complies with the Unique Change of State constraint in the
sense that it further specifies the result designated by the adverb out. Therefore, we can have conceptually feasible
combinations of results provided that they are compatible. The same holds true for the intransitive resultative
construction which employs a combination between an adverb and a prepositional phrase This thinking blossomed
out in Buddhism’s greatest contribution to mankind, namely the concept of mettà [Skt. maitrã] or universal loving
kindness (Sketch engine doc#940191). This is a metaphorical expression that makes use of the basic emergence
meaning of the verb blossom. The emergence of a flower out of the stem of a plant (in the source domain of
the metaphor) is mapped onto the emergence of an ideological precept out of an ideological movement. The
preposition in does not code any motion but it figuratively expresses a state (kindness). Its use is licensed by the
low-level metaphor STATES ARE LOCATIONS.
5. CONCLUSIONS
In this article it has been shown that seven of Levin’s (1993) entity-specific change-of-state verbs, namely
bloom, blossom, flower, germinate, sprout, swell, and blister, are conceptually similar in the sense that they refer
to an increase in size, volume, or intensity of a particular entity. They also select the A>A’ resultative schema that
indicates the acquisition of a new property (e.g. Gorse blossomed gold on magnesium limestone embankments;
COCA 1994). Furthermore, the intransitive resultative construction can express either a simple or a compound
result. The former can be lexicalized by an AP (e.g. […] a series of concentric burns blistered black […]) or a PP
(e.g. […] this encounter blossomed into the most intense relationship of Goldman’s life). The latter can be encoded
by means of a combination between an adverb and an AP (e.g. Their throats would swell out big […], where the
AP big further specifies the result denoted by the adverb out) or a combination between an adverb and a PP (e.g.
[…] she had blossomed out into a lovely womanhood […]). It has also been noted that on some occasions the
AP expressing a change of color can be syntactically separated from the verb by means of prepositions, such
RLyLA Vol. 09 (2014), 97-105 | 103
Andreea Rosca
as in or into. A sentence like The flower bloomed in motley red is liable to a part-affectedness interpretation in
the sense that the color that covers the surface of the flower displays different hues. Finally, the subsumption
processes between these verbs and the prepositional intransitive resultative construction observe the Internal
Variable Conditioning constraint formulated by the LCM, according to which the world-knowledge associated to
an internal predicate variable restricts the nature of both the predicate and its constructional arguments.
references
Boas, H. C. (2008). “Resolving Form-Meaning Discrepancies in Construction Grammar”, in J. Leino (ed.)
Constructional reorganization. Amsterdam and Philadelphia: John Benjamins, pp. 11-36. doi:10.1075/
cal.5.02boa
Broccias, C. (2004). “The Cognitive Basis of Adjectival and Adverbial Resultative Constructions”. Annual Review
of Cognitive Linguistics, 2: 103-126. doi:10.1075/arcl.2.04bro
Butler, C. S. (2009). “The Lexical Constructional Model: Genesis, strengths and challenges”, in C. S. Butler and J.
Martín Arista (eds.) Deconstructing Constructions. Amsterdam and Philadelphia: John Benjamins, pp. 117152. doi:10.1075/slcs.107.07the
Faber, P. B. and R. Mairal Usón (1999). Constructing a lexicon of English Verbs. Berlin and New York: Mouton de
Gruyter. doi:10.1515/9783110800623
Fodor, J. A. (1970). “Three Reasons for Not Deriving Kill from Cause to Die”. Linguistic Inquiry, 1: 429-438.
Goldberg, A. (1991). “It can’t go down the chimney up: Paths and the English Resultative”, in L. A. Sutton, C.
Johnson and R. Shields (eds.) Proceedings of the seventeeth annual meeting of the Berkeley Linguistics
Society. Berkeley, California: Berkeley Linguistics Society, pp. 368-378.
Goldberg, A. (2006). Constructions at work: the nature of generalization in language. New York: Oxford University
Press.
Haiman, J. (1980). “Dictionaries and Encyclopedias”. Lingua, 50: 329-357. doi:10.1016/0024-3841(80)90089-3
Iwata, S. (2005). “The Role of Verb Meaning in Locative Alternations”, in M. Fried and H. C. Boas (eds.) Grammatical
Constructions: Back to the Roots. Amsterdam and Philadelphia: John Benjamins, 101-118. doi:10.1075/
cal.4.07iwa
Lakoff, G. (1987). Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago: The
University of Chicago Press. doi:10.7208/chicago/9780226471013.001.0001
Lakoff, G. and M. Turner (1989). More than Cool Reason: A Field Guide to Poetic Metaphor. Chicago: University of
Chicago Press. doi:10.7208/chicago/9780226470986.001.0001
Lakoff, G. and M. Johnson (1980). Metaphors We Live By. Chicago: The University of Chicago Press.
Levin, B. (1993). English Verb Classes and Alternations. A Preliminary Investigation. Chicago and London: University
of Chicago Press.
Levin, B. (2006). English Object Alternations: A Unified Account. Unpublished ms., Stanford University: Stanford,
CA.
Luzondo, A. (2011). English resultative constructions in the Lexical Constructional Model: implications for
constructional modelling within a lexical conceptual knowledge base. Doctoral dissertation, University of
La Rioja.
Mairal, R. and F. J. Ruiz de Mendoza (2008). “New challenges for lexical representation within the LexicalConstructional Model”. Revista Canaria de Estudios Ingleses, 57: 137-158.
Mairal, R. and F. J. Ruiz de Mendoza (2009). “Levels of description and explanation in meaning construction”, in
C. S. Butler and J. Martín Arista (eds.) Deconstructing Constructions. Amsterdam and Philadelphia: John
Benjamins, pp. 153-198. doi:10.1075/slcs.107.08lev
Martín Mingorance, L. (1998). El modelo lexemático-funcional: El legado lingüístico de Leocadio Martín Mingorance.
Ed. A. Marín Rubiales. Granada: University of Granada.
Nemoto, N. (2005). “Verbal Polysemy and Frame Semantics in Construction Grammar: Some observations on
the locative alternation”, in M. Fried and H. C. Boas (eds.) Grammatical Constructions – Back to the Roots.
Amsterdam and Philadelphia: John Benjamins, pp. 119-136. doi:10.1075/cal.4.08nem
Östman, J.-O. and M. Fried (2004). “Historical and intellectual background of Construction grammar”, in M.
Fried and J.-O. Östman (eds.) Construction Grammar in a cross-language perspective. Amsterdam and
Philadelphia: John Benjamins, pp. 1-10. doi:10.1075/cal.2.01ost
Ruiz de Mendoza, F. J. (2008). “Cross-linguistic analysis, second language teaching and cognitive semantics: the
case of Spanish diminutives and reflexive constructions”, in S. De Knop and T. De Rycker (eds.) Cognitive
Approaches to Pedagogical Grammar. Berlin and New York: Mouton de Gruyter, pp. 121-152.
RLyLA Vol. 09 (2014), 97-105 | 104
Andreea Rosca
Ruiz de Mendoza, F. J. (2011). “Metonymy and cognitive operations”, in R. Benczes, A. Barcelona and F. J. Ruiz de
Mendoza (eds.) Defining Metonymy in Cognitive Linguistics: Towards a Consensus View. Amsterdam and
Philadelphia: John Benjamins, pp. 103-124. doi:10.1075/hcp.28.06rui
Ruiz de Mendoza, F. J. (2013). “Meaning construction, meaning interpretation and formal expression in the Lexical
Constructional Model”, in Elke Diedrichsen and Brian Nolan (eds.) Linking constructions into functional
linguistics: The role of constructions in RRG grammars (Studies in Language Series). Amsterdam and
Philadelphia: John Benjamins; vol. in prep.
Ruiz de Mendoza, F. J. and A. Luzondo (2013). “Lexical-constructional subsumption in resultative constructions
in English”, in M. Brdar, M. Zic, I. Raffaelli, M.-M. Stanojevic and N. Tudjman Vukovic (eds.) Cognitive
Linguistics. Between Universality and Variation. Cambridge Scholars Publishing; vol. in prep.
Ruiz de Mendoza, F. J. and R. Mairal (2007). “High-level metaphor and metonymy in meaning construction”, in
G. Radden, K. M. Köpcke, T. Berg and P. Siemund (eds.) Aspects of Meaning Construction in Lexicon and
Grammar. Amsterdam and Philadelphia: John Benjamins, pp. 33-49.
Ruiz de Mendoza, F. J. and R. Mairal (2008). “Levels of description and constraining factors in meaning construction:
An introduction to the Lexical Constructional Model”. Folia Linguistica, 42/2: 355-400.
Ruiz de Mendoza, F. J. and R. Mairal (2011). “Constraints on syntactic alternation: lexical-constructional
subsumption in the Lexical-Constructional Model”, in P. Guerrero (ed.) Morphosyntactic Alternations in
English. Functional and Cognitive Perspectives. London, UK/Oakville, CT: Equinox, pp. 62-82.
Ruiz de Mendoza, F. J. and F. Santibáñez (2003). “Content and formal cognitive operations in construing meaning”.
Italian Journal of Linguistics, 15: 293-320.
Van Valin, R. (2005). The syntax-semantics-pragmatics interface: an introduction to Role and Reference Grammar.
Cambridge: Cambridge University Press. doi:10.1017/CBO9780511610578
Wright, S. K. (2002). “Transitivity and Change of State Verbs”, in J. Larson and M. Paster (eds.) Proceedings of the
twenty-eighth annual meeting of the Berkeley Linguistics Society. Berkeley, California: Berkeley Linguistics
Society, pp. 339-350.
RLyLA Vol. 09 (2014), 97-105 | 105
Vol. 9 año 2014, 106-116
EISSN 1886-6298
Split and unified functions in
the formation of Old English nouns and adjectives1
Raquel Vea Escarza
Abstract: This journal article draws a distinction between the split and unified functions obtaining in the formation of Old English
nouns and adjectives. The starting point of the discussion is an enlarged inventory of lexical functions that draw on MeaningText Theory and structural-functional grammars and explain the change of meaning caused by prefixation and suffixation in Old
English. The extended inventory of lexical functions consists of 33 functions and has been applied to ca. 7,500 affixed nouns
and adjectives extracted from the lexical database Nerthus (www.nerthusproject.com). The distinction between split and unified
functions, in such a way that the former can be realized by both prefixes and suffixes and the latter by either prefixation or
suffixation, allows for some generalizations. Firstly, the analysis proves that there are more functions involved in prefixation than
in suffixation. Secondly, prefixation is meaning oriented while suffixation is class oriented.
Keywords: lexical functions, structural-functional grammar, affixation, nouns, adjectives, Old English.
1. INTRODUCTION
Recent research in the lexical semantics of Old English pays more attention to morphological than semantic
aspects. Kastovsky (2006) identifies a typological change in the morphology of Old English whereby variable
bases of derivation have been replaced by invariable bases. Kastovsky distinguishes two steps in this evolution:
from root-formation to stem-formation and from stem-formation to word-formation. The work by Haselow (2011)
follows in the same track, although he is more concerned with derivational morphology than Kastovsky, who
focuses on inflectional phenomena such as ablaut in strong verbs or the simplification of inflection and explains
their consequences to derivational morphology from the perspective of inflection. Haselow (2011) addresses
the question of productivity in noun formation and finds some rising analytic tendencies that can be seen as a
consequence of the change to invariable base morphology which, being unable to modify the root to produce a
stem or a stem to produce a word, has to add affixes (thus analytic) to bases of derivation that enjoy the status
of words.2 Martín Arista (2008, 2009, 2011c, 2012b) puts forward a framework of functional morphology that is
inspired in the layering of functional grammars and, consequently, focuses on the points of contact between
morphology and syntax. This author has applied different aspects of the morphological framework of the Layered
Structure of the Word to the derivational morphology of Old English (Martín Arista 2010a, 2010b, 2011a, 2012a) as
well as the structure of the Old English lexicon in general (Martín Arista 2011b, 2013).
The works just cited, in spite of their valuable contribution to the study of derivational morphology and the
structure of the Old English lexicon, do not engage in the changes of meaning caused by the derivational processes
of word-formation. Indeed, Kastovsky (1992) provides lists of affixes, derivatives and processes or explains the
relationship between inflection and derivation in terms of the status of morphological bases (Kastovsky 2006).
Martín Arista (2008, 2009, 2011c, 2012b) explains the similarities between syntactic and morphological units by
means of the concept of layering (or increasing semantic-syntactic complexity) and devises a syntactic model of
morphology that uses templates, positions and functions at word level. Such functions, however, express only
syntactic relations of constituency or dependency and have little to do with meaning.3 A line of research more
1
2
3
This research has been funded through the project FFI2001-29532, which is gratefully acknowledged.
Trips (2009) looks at the same series of Old English noun-forming suffixes as Haselow (2011) although she puts more emphasis on the question of productivity.
Cf. Martín Arista and Cortés Rodríguez (fc.).
| 106
Raquel Vea Escarza
Split and unified functions in the formation of Old English nouns and adjectives
concerned with meaning is the one in the semantic primes of Old English pursued by Martín Arista and Martín
de la Rosa (2006), who apply the methodology and inventory of primes of the Natural Semantic Metalanguage
Research Programme (Goddard and Wierzbicka 2002; Wierzbicka 2002) to the substantives, determiners and
quantifiers of Old English. In subsequent publications, de la Cruz Cabanillas deals with descriptors, Guarddon
Anelo (2009a, 2009b) with compound adpositions and Mateo Mendaza (2013) with TOUCH. These works, however,
revolve around the basic lexicon or, when they take derivatives into account, the parameter of word-formation is a
secondary one (Mateo Mendaza 2013) or is not considered.
Another relevant aspect for a review of previous research in the area of Old English lexical semantics is the
overall syntagmatic or paradigmatic approach to lexical items and lexical relations in general and to those involved
in derivational processes in particular. Kastovsky (1992, 2006) opts for an exclusively syntagmatic approach
because he does not make generalizations applicable to sets of affixes or sets of meanings. Martín Arista (2011b,
2013) makes two types of generalizations characteristic of the paradigmatic axis of analysis. Firstly, he devises
derivational paradigms (comparable to word-families but with an explicit and principled hierarchical structure)
that have a dynamic part (the derivational processes that produce all the derivatives in the paradigm) and a static
one (the output of such processes). Secondly, he distinguishes two lexical layers in Old English that depend
on the formal and semantic transparency and relative productivity of derivational processes and introduces the
parameter of recursivity (or derivation of previously derived words), in such a way that the layer of zero derivation
is subservient to affixation and affixation is subservient to recursive derivation. In the remainder of this work the
line is taken that the paradigmatic approach adopted by Martín Arista (2011b, 2013) can be enriched by identifying
lexical functions in word-formation as a means of making generalizations that cut across derivational paradigms
and tightening up the lexical relations that hold among bases.
On the grounds of the preceding review, it seems necessary to adopt a paradigmatic perspective on wordformation and, moreover, to deal with its semantic side more clearly than previous research. As a contribution to
these lines of research, this article aims at explaining the change of meaning caused by the derivational processes
of Old English that turn out nouns and adjectives. More specifically, the concept of split vs. unified funcion, based
on positional realization, is coined in order to determine whether or not prefixation and suffixation change in a
different way the meaning of the base of derivation. The theoretical foundations are provided by the structuralfunctional theory of linguistics.
The remainder of the article is organized as follows. Section 2 discusses the methodological aspects of
the research, including pre-theoretical questions, data and units of analysis. Section 3 develops an extended
framework of lexical functions that draws on Pounder (2000) and is enlarged by following the structural-functional
theories of language. Section 4 applies the extended framework of lexical functions to the affixed (prefixed and
suffixed) nouns and adjectives of Old English and puts forward the concepts of split and unified lexical function.
After the discussion of the results of the analysis from the perspective of these concepts, the conclusions are
presented in section 5.
2. METHODOLOGY
Beginning with the question of explanation, explanation in functional grammars (Foley and Van Valin 1984;
Dik 1997a, 1997b; Van Valin 1997; Van Valin 2005; Hengeveld and Mackenzie 2008) is mostly couched in terms
of functions. Functions, as relations between linguistic units that capture the systematic aspects of language and
thus allow generalizations, are at the core of functional-structural grammars like the ones cited above. As in these
works, explanation in this article is sought through functions.4
The question of the change of meaning is addressed in an exclusively synchronic fashion. This research,
therefore, does not belong in etymology. Rather, it centres on the meaning relations that hold in an isolated stage
of the English language like Old English. This does not mean that the importance of the origin and the subsequent
historical evolution of a given word are underestimated but the simultaneous analysis of all synchronic and
diachronic facts is incompatible with the exhaustive analysis carried out in this work, which has scope over all
affixed nouns and adjectives.5
The study of meaning in a historical language is more dependent on dictionary meaning than is meaning in
a living language because there are no native speakers and, consequently, no intuitions or judgements about
their linguistic abilities are available. Moreover, the limits of the knowledge of a historical language is completely
constrained by the surviving written records, which are unlikely to reflect all and every detail of lexical organization.
4
5
See Butler (2003a, 2003b) on functional relations and the organization of the major structural-functional theories of grammar.
Martín Arista (2013) proposes the concept of derivational depth, which combines some of the synchronic and diachronic aspects of the formation of a lexical paradigm.
RLyLA Vol. 09 (2014), 106-116 | 107
Raquel Vea Escarza
This means that, in this article, meaning is used with the value of dictionary meaning. The unit of analysis where
meaning is sought is the word rather than the sentence of the utterance.6
As for the derivational processes, this work assumes a strictly gradual application of the processes of wordformation, in such a way that a maximum of one affix is attached by any given process.7 This has two consequences.
If a maximum of one morpheme is attached by derivational step, zero derivation is considered a relevant process
for Old English word-formation. The second is that word-formation requires explicit or implicit morphemes (bound
forms), which excludes free forms as inputs to derivational processes. Regarding the derivational processes of Old
English, it is also necessary to remark that no assessment of productivity is done in the research that follows. This
means that no difference is made between productive and non-productive processes (or even between more and
less productive processes) provided that the output of the process is still formally analysable.
In prefixation, the following prefixes can be distinguised. Textual realizations and variants are given between
brackets:
(1) Ā- (ā-), Ǣ- (ǣ-), ÆFTER- (æfter-), ÆT- (æt-), AND- (am-, an-, and-), ANTE- (ante-), ARCE- (arce-), BE- (bi-,
bī), EALL- (æl-, al-, eall-), ED- (ed-), EL- (æl-, el-), FOR- (for-, fōr-, fore-), FORE- (for-, fore-, fōre-), FORÐ(forð-), FRAM- (fram-), FRĒA- (frēa-), FUL- (ful-, full-), GĒAN- (gean-, gēan-), HEALF- (healf-), IN- (in-,
inn-), MID- (med-, mid-), Ō- (ō-), OF- (æf-, of-), OFER- (ofer-), ON- (on-), OR- (ō-, or-), SĀM- (sam-, sām-),
SIN- (sin-, sine-), SUB- (sub-), TŌ- (tō-), TWI- (twi-), ÐRI- (ðri-, ðry-), ÐURH- (ðurh-), UN- (and-, on-, un-),
UNDER- (under-), ŪP- (up-, ūp-), ŪT- (ūt-, ūð-),WAN- (wan-), WIÐ- (wið-), WIÐER- (wiðer-), YMB- (ymb-,
ymbe-).
Suffixes with their textual realizations and variants are displayed in (2). (2a) presents the nominal suffixes and
(2b) the adjectival suffixes distinguished in the analysis:
(2) a. -BORA (-bior, -bora), -DŌM (-dōm), -ED (-ad), -EL (-el, -eld, -ele, -elle, -il, -l, -la, -le, -ll, -lle, -ol), -ELS
(-els, -ls), -EN (-en, -n), -END (-d, -en, -end, -ende, -iend, -liend, -nd), -ERE (-e, -er, -era, -ere, -igere,
-lere, -lēre, -re), -ESSE (-esse), -ESTRE (-estre, -istre, -stre, -ystre), -ETT (-et, -eta, -ett, -t, -tt), -FUL
(-ful), -HĀD (-hād), -ICGE (-ecge, -icge, -ige), -IG (-ig), -INCEL (-incel), -ING (-ing, -unga, -inga), -LING
(-ling), -NES (-enes, -es, -nes, -ness, -nis, -nys, -nyss, -s), -RǢDEN (-rǣden), -SCIPE (-scipe, -scype),
-SUM (-sum), -ð (-að, -d, -ed, -ot, -oð, -oða, -t, -ð, -ða, -ðe, -ðo, -ðu, -uð), -UNG (-ng, -ung), -WIST
(-wist).
b. -BǢRE (-bǣre), -CUND (-cund), -ED (-ade, -ed, -ede, -od, -ode, -te, -ud), -EL (-el, -ol, -ul), -EN (-en),
-END (-end, -igend), ENDE (-ende, -iende), -ERNE (-ern, -erne), -FÆST (-fæst), -FEALD (-feald), -FUL
(-ful), -IC (-ic), -IG (-ig, -ige), -IHT (-eht, -ehte, -iht, -ihte), -ING (-ing), -ISC (-isc), -LĒAS (-lēas), -LIC
(-lic), -OR (-or), -SUM (-sum), -WEARD (-weard), -WELLE (-welle), WENDE (-wende).
The data comprise 7,588 complex words. By category, they can be broken down into 4,370 affixed nouns and
3,218 affixed adjectives. By process, they consist of 2,001 prefixed words and 5,587 suffixed ones. They have been
retrieved from the lexical database of Old English Nerthus (www.nerthusproject.com), which is based on An AngloSaxon Dictionary (Bosworth and Toller 1973), The Student´s Dictionary of Anglo-Saxon (Sweet 1976) and, above
all, A Concise Anglo-Saxon Dictionary (Clark Hall 1996). On specific questions relating to meaning definitions, The
Dictionary of Old English (Healey 2008) has been consulted, while some decisions on morphological relatedness
have been made on the basis of the data provided by The Dictionary of Old English Corpus (Healey et al. 2004).
Finally, on some minor etymological questions, the standard dictionaries of Germanic and Old English etymology
have been looked up, including Holthausen (1963), Pokorny (1989) and Orel (2003), as well as Seebold (1970) and
Heidermanns (1993).
3. THE LEXICAL FUNCTIONS OF OLD ENGLISH WORD-FORMATION
Pounder (2000) carries out a paradigmatic analysis of the formation of German adjectives in which the
relationship holding between the meaning of a derivative and its base is accounted for in terms of a lexical
function. For the concept of lexical function, Pounder (2000) draws on Meaning-Text Theory (Mel’čuk 1989, 1996,
2006), which goes beyond semantic relations in order to identify lexical functions between words. This theory was
firstly applied to Russian and French and devised more than 60 lexical functions that appear in the Explanatory
Combinatorial Dictionary (Mel’čuk, Clas, and Arbatchewsky-Jumarie 1984-99). Pounder draws a basic distinction
6
7
See Kastovsky (1992) and Lass (1994) on the issues that arise in the lexical analysis of historical languages.
See Lacalle Palacios (2011) on the limits of gradual affixation in Old English. See also Torre Alonso (2009, 2011a, 2011b, 2012) on recursivity.
RLyLA Vol. 09 (2014), 106-116 | 108
Raquel Vea Escarza
between primary and secondary functions, so that secondary functions modify word-formation meanings rather
than constituting word-formation meaning on their own (Pounder 2000: 109). In this view, a function typically used
in combination with primary functions is the pejorative. The Old English data, however, clearly indicate that the
distinction between primary and secondary functions is not relevant for the analysis of this language. It has been
found, in the first place, that some functions considered secondary by Pounder (2000) apply to underived lexical
items. This is the case with the following functions:
(3)PEJ(orative)(‘X’): mislār ‘ill teaching’ (< LĀR ‘teaching’)
INTENS(ifier)(‘X’): eallgrēne ‘all green’ (< GRĒNE 18 ‘green’)
AUGM(entative)(‘X’): fullmægen ‘great power’ (< MÆGEN ‘power’)
The Old English data also evidence that Pounder´s (2000) primary functions apply to derived lexical items that
already hold a lexical function. This happens in combinations of functions that involve REL(ative)(‘X’), EX(origin)
(‘X’), DIST(ributive)(‘X’) and I(dentity)(‘X’), as is shown in (4a), (4b), (4c) and (4d) respectively:
(4) a. WITHPROP(‘X’) < REL(‘X’)
wlitigfæst ‘of enduring beauty’ < wlitig ‘radiant’ (< WLITE ‘brightness’)
b. WITHPROP(‘X’) < EX(‘X’)
fȳrenful ‘fiery’ < fȳren ‘of fire’(< FȲR ‘fire’)
c. REL(‘X’) < DIST(‘X’)
ðrifealdlic ‘triple’ < ðrifeald ‘three-fold’(< ÐRĪE ‘three’)
d. REL(‘X’) < I(‘X’)
sāriglic ‘sad’ < sārig ‘sore, grieved’ (< NĒAT ‘sore’)
Although the distinction between primary and secondary functions is not made in the present study, a
number of lexical functions have been adopted from Pounder´s (2000) work directly, including REL(‘X’), EX(‘X’),
LIKE(similarity)(‘X’), DIM(inutive)(‘X’), I(‘X’), PEJ(‘X’), INTENS(‘X’) and DIST(‘X’). They are illustrated in (5):
(5)DIM(‘STĀN’): stānincel ‘little stone’
DIST(‘HUND 1’): hundwelle ‘a hundred-fold’
EX(‘LĀR’): lārdōm ‘teaching, instruction’
I(‘WEDD’): anwedd ‘pledge’
INTENS(‘HWĪT 1’): eallhwīt ‘entirely of white’
LIKE(‘WĪN’): wīnlic ‘like wine’
PEJ(‘TĪD’): mistīd ‘evil time’
REL(‘HOF’): hoflic ‘pertaining to a court’
A study in Old English word-formation requires additional functions, whose motivation has been sought in the
structural-functional tradition of linguistics. In a nutshell, Functional Grammar (Dik 1997a, 1997b) has provided the
typology of entities and Role and Reference Grammar (Van Valin and LaPolla 1997; Van Valin 2005) has contributed
the roles associated with the semantic macroroles ACTOR and UNDERGOER. The latest framework of Functional
Grammar, Functional Discourse Grammar (Hengeveld and Mackenzie 2008), has supplied additional semantic
categories, while the layered framework consisting of semantically defined units is, to a certain extent, common to
all three theories (Butler 2003a, 2003b).
Dik’s (1997b) taxonomy of basic semantic categories has provided inspiration for the function PRED(icative)
(‘X’), which indicates deverbal nominalizations that embed a predication. In turn, the semantic category Individual
is the basis of the lexical function COM(itative)(‘X’), which coincides with the semantic role defined in Functional
Grammar.
Functional Discourse Grammar (Hengeveld and Mackenzie 2008) has completed Lyon’s (1995) semantic
taxonomy to account for, among others, Time, Location and Quantity. For instance, the semantic category Location
motivates the function LOC(ative)(‘X’). In turn, the semantic category Time motivates two lexical functions, one
literal, the other one figurative, namely TEMP(oral)(‘X’) and LIKE(TEMP(oral)(‘X’)). Finally, the semantic category
Quantity has motivated the function MASS(ive)(‘X’), which makes reference to uncountable amounts.
With the addition of the lexical functions inspired by structural-functional theories of language, it can be the
case that a certain function further specifies one proposed by Pounder (2000). For instance, WITH(‘X’) has been
divided into WITHENT(ity)(‘X’) and WITHPROP(erty)(‘X’) in the case of adjectives, which give rise to their nominal
8
As in the lexical database Nerthus, numbered predicates are used to indicate that homophones belong to different lexical categories or morphological classes. For
example, grēne 1 is an adjective and grēne 2 a noun.
RLyLA Vol. 09 (2014), 106-116 | 109
Raquel Vea Escarza
counterparts, namely ENT(ity)(‘X’) and PROP(erty)(‘X’). It can also be the case that a pair of new functions are
related to a certain function put forward by Pounder. For example, the functions MAGN(ifier)(‘X’) and MIN(imiser)
(‘X’) are related to Pounder´s AUGM(‘X’) and DIM(‘X’) but the labels MAGN and MIN have been preferred to make
reference to rank, although DIM(‘X’) is used after Pounder for diminutives. It remains to say that figurative meaning
in this framework results from the function of similarity and the corresponding non-figurative meaning, as in
LIKE(LOC(ative)(‘X’)) ‘figurative locative’ as applying in instances like forðāurnen ‘elapsed’ (< FORÐĀGAN ‘to pass
away’)V.
Although functional theories of language have been the main inspiration for extending the inventory of lexical
functions, other sources must be acknowledged. The typology of lexical negation is based on Martín Arista (2010b)
and relies on lexical category: privation requires a nominal base, counterfactuality a verbal base and opposition
an adjectival base. Thus, Pounder´s (2000) function NEG(ation)(‘X’) has been broken down into three functions of a
more specific nature: PRIV(ative)(‘X’), OPP(ositive)(‘X’) and COUNTFACT(ual)(‘X’). Finally, the functions ABST(ract)
(‘X’), PART(itive)(‘X’) and MIT(igator)(‘X’) draw on Lexeme-Morpheme Base Morphology (Beard 1995; Beard and
Volpe 2005).
Although there is a FEM(inine)(‘X’) lexical function, no masculine or neuter counterparts have been defined.
This is so for two reasons. Firstly, there is a significant number of pairs feminine-masculine that share a lexemic
basis, but pairs comprising the masculine and the neuter or the feminine and the neuter are hard to find. Secondly,
when the suffixes of Old English nouns are considered (specifically -nes, and -ung, which form feminines),
the feminine turns out to be the marked gender of the first argument and the unmarked gender of the second
argument. Conversely, the masculine is the unmarked gender of the first argument and the marked gender of the
second. Moreover, feminine derivatives tend to be abstract whereas masculine derivatives show a tendency to be
concrete. These reasons can be put in another way by remarking that the first argument has been broken down
into five functions (AG(ent)(‘X’), EFF(ector)(‘X’), PAT(ient)(‘X’), STA(tive)(‘X’) and FEM(‘X’)), which, in practice adds an
additional gender feature to agents, effectors and patients). This is consistent with the prototypical first argument
of a predication, which is conscious, animate and human. On the contrary, the second argument is underspecified
by comparison (patient and predicative).
With these premises, the set of lexical functions relevant for Old English affixal nouns and adjectives includes:
ABST(‘X’) ‘abstract’ holds in nouns with nominal base whose meaning becomes less concrete, as
in winescipe ‘friendship’ (< WINE 1 ‘friend’)N
AG(‘X’) ‘agent’ holds in nouns with verbal base that qualify as prototypical agents, as in cwylmend
‘destroyer’ (< CWYLMIAN ‘to torment’)V
AUGM(‘X’) ‘augmentative’ holds in nouns whose referent becomes increased, as in oferwist
‘gluttony’ (< MIHT 1 ‘might’)N
COM(‘X’) ‘comitative’ holds in nouns that involve company or association, as in samwist ‘living
together’ (< WIST ‘provision’)N
COUNTFACT(‘X’) ‘counterfactual’ holds in adjectives with verbal bases that convey a meaning of
reversal of the verbal base, as in unscirped ‘unclothed’ (< SCIERPAN ‘to clothe’)V
DIM(‘X’) ‘diminutive’ holds in nouns which denote a reduced size of the referent, as in tūnincel
‘small property’ (< TŪN ‘enclosure’)N
DIST(‘X’) ‘distributive’ holds in nouns and adjectives that denote the scaling of the referent by an
amount or number, as in twiwǣg ‘junction of two roads’ (< WEG ‘road’)N
EFF(‘X’) ‘effector’ holds in nouns with verbal base. It denotes a non-prototypical agent, as in
andettere ‘one who confeses’ (< (GE)ANDETTAN ‘to confess’)V
ENT(‘X’) ‘entity’ holds in nouns that refer to an entity, thus concrete and identifiable in space and
time, as in cæfing ‘hair-ornament’ (< CÆFIAN ‘to embroider’)V
EX(‘X’) ‘origin’ holds in both nouns and adjectives. This function defines ‘X’ as the origin, as in
Centisc ‘Kentish’ (< CENT ‘Kent’)N
FEM(‘X’) ‘feminine’ holds in nouns that have a masculine counterpart, as in lybbestre ‘sorceress’
(< LYBB ‘poison’)N
I(‘X’) ‘identity’ holds in nouns and adjectives when the base of derivation and the derivative are
partial synonyms, as in grimlic ‘fierce’ (< GRIMM ‘fierce’)Adj
INTENS(‘X’) ‘intensifier’ holds in adjectival predicates whose properties are intensified with respect
to the base, as in eallbeorht ‘all bright’ (< BEORHT 1 ‘bright’)Adj
RLyLA Vol. 09 (2014), 106-116 | 110
Raquel Vea Escarza
LIKE(‘X’) ‘similarity’ holds in adjectives that take a characteristic of ‘X’ to establish a comparison,
as in āncorlic ‘like a hermit’ (< ĀNCOR ‘hermit’)N
LIKE(LOC(‘X’)) ‘figurative locative’ holds in nouns and adjectives to convey a figurative sense of
location, as in wiðerwyrd ‘contrary’ (< (GE)WYRD ‘fate’)N
LIKE(TEMP(‘X’)) ‘figurative temporal’ holds in nouns. This function involves a figurative sense of
time, as in foreðanc ‘forethought’ (< (GE)ÐANC ‘thought’)N
LOC(‘X’) ‘locative’ holds in nouns and adjectives and expresses concrete location, as in inwund
‘internal wound’ (< WUND 1 ‘wound’)N
MAGN(‘X’) ‘magnifier’ holds in nouns to denote a higher rank than that of the base, as in framrinc
‘chief’ (< RINC ‘man’)N
MASS(‘X’) ‘mass’ holds in nouns. This function is used to denote a pluralization or collectivization
of the referent of the base, as in scræfen ‘place with caves’ (< (GE)SCRÆF ‘cave’)N
MIN(‘X’) ‘minimiser’ holds in nouns and denotes a lower rank than the referent, as in undercyning
‘under-king’ (< CYNING ‘king’)N
MIT(‘X’) ‘mitigator’ holds in adjectives whose characteristics become diminished, as in healfrēad
‘reddish’ (< RĒAD ‘read’)Adj
OPP(‘X’) ‘oppositive’ holds in nouns and adjectives to convey the meaning of the opposite of the
entity or property expressed by the base, as in uncynn ‘unsuitable’ (< CYNN 2 ‘suitable’)N
PART(‘X’) ‘partitive’ holds in nouns to indicate that a part of a larger amount is referred to, as in
healffers ‘hemistich’ (< FERS ‘verse’)N
PAT(‘X’) ‘patient’ holds in nouns that convey the meaning of the patient of a stative verb, as in
swefecere ‘sleeper’ (< SWEFAN ‘to sleep’)V
PEJ(‘X’) ‘pejorative’ holds in nouns and in adjectives when the meaning of the base is evaluated
negatively, as in mishworfen ‘perverted’ (< (GE)HWEORFAN ‘to turn’)V
PRED(‘X’) ‘predicative’ holds in nouns and expresses the objective type of deverbal predications,
as in sealtnes ‘saltness’ (< SEALTAN ‘to salt’)V
PRIV(‘X’) ‘privative’ holds in nouns and adjectives and denotes the absence of a property or entity,
as in egelēas ‘fearless’ (< EGE ‘fear’)N
PROP(‘X’) ‘property’ holds in nouns and expresses an abstract property, as in wearmnes ‘warmth’
(< WEARM ‘warm’)Adj
REL(‘X’) ‘relative’ holds in adjectives and assigns a reference relation to the meaning of the base,
as in bōclic ‘of or belonging to a book’ (< BŌC ‘book’)N
STA(‘X’) ‘stative’ holds in adjectives and expresses a property derived from a verbal base, as in
hospul ‘despised’ (< HOSPAN ‘to reproach’)V
TEMP(‘X’) ‘temporal’ holds in nouns and adjectives and expresses a temporal reference, as in
midsumer ‘midsummer’ (< SUMOR ‘summer’)N
WITHENT(‘X’) ‘with entity’ holds in adjectives to define the possession of an entity, which is
concrete, as in loccod ‘hairy’ (< LOCC 1 ‘lock, hair’)N
WITHPROP(‘X’) ‘with property’ holds in adjectives that enjoy a property, which is abstract, as in
ātorbǣre1 ‘poisonous’ (< ĀTOR ‘poison’)N
Figure 1: The extended framework of lexical functions.
In spite of the applicability of Pounder´s (2000) lexical functions to this study, which may be explained as a
consequence of the Germanic background common to English and German, it has been necessary to resort to
proposals made mainly by Functional Grammar, Functional Discourse Grammar and Role and Reference Grammar
because some functions defined by Pounder have little applicability to the data, like SING(‘X’), and because some
lexical functions, especially in nouns, cannot be accounted for by Pounder´s inventory. It has to be borne in mind
in this respect that it can be the case that a certain function drawn from another framework further specifies one
proposed by Pounder. For instance, WITH(‘X’) has been broken down into WITHENT(‘X’) and WITHPROP(‘X’). It
can also be the case that a pair of new functions are related to a certain function put forward by Pounder. For
example, the functions MAGN(‘X’) and MIN(‘X’) are related to Pounder´s AUGM(‘X’) and DIM(‘X’) but the labels
MAGN and MIN have been preferred to make reference to rank, although, as it has just been remarked, DIM(‘X’)
is used after Pounder for diminutives.
RLyLA Vol. 09 (2014), 106-116 | 111
Raquel Vea Escarza
4. SPLIT vs. UNIFIED FUNCTIONS
To recapitulate, the preceding sections have reviewed the relevant literature on the lexical semantics of Old
English and presented an extended framework of lexical functions applicable to the analysis of the affixed nouns
and adjectives of Old English. This section discusses the results of the analysis and puts forward the concepts of
split and unified lexical function. These concepts allow to draw conclusions on the contribution of meaning made
by prefixation and suffixation.
Figure 2 summarizes the results of the analysis that has been carried out.
Category
Lexical Function
Affixes
AdjectiveCOUNTFACT(‘X’)unAdjectiveDIST(‘X’)
-feald, twi-, ðri-, -wele
AdjectiveEX(‘X’)
ā-, -cund, -ed, -en, -erne, -ic, -isc
AdjectiveI(‘X’)
ǣ-, æt-, and-, eall-, -el, for-, ful-, -ig, -lic, on-, sin-, tō-, unAdjectiveINTENS(‘X’)
eall-, for-, forð-, frēa-, ful-, ofer-, on-, sin-, ðurh-, ūtAdjectiveLIKE(‘X’)
-ed, -en, -ig, -iht, -lic
AdjectiveLIKE(LOC(‘X’)) fore-, forð-, fram-, in-, of-, under-, ūt-, -weard, wiðerAdjectiveLOC(‘X’)
and-, forð-, fram-, in-, of-, ofer-, under-, ūt-, -weard
AdjectiveMIT(‘X’)
healf-, mid-, sāmAdjectiveOPP(‘X’)
am-, and-, in-, unAdjectivePEJ(‘X’)
misAdjectivePRIV(‘X’)
ǣ-, æt-, and-, -lēas, ofer-, or-, wanAdjectiveREL(‘X’)
-ig, -lic
AdjectiveSTA(‘X’)
-el, -end, -ende, -ing, -or
AdjectiveTEMP(‘X’)
æfter-, for-, foreAdjectiveWITHENT(‘X’)
-ade, and-, -bǣre, -ed, -eht, -ende, -fæst, -ful, -iht, -welle
AdjectiveWITHPROP(‘X’) -bǣre, -ed, -ede, -en, -ende, -fæst, for-, -ful, -iht, -or, -sum, -welle
NounAG(‘X’)
-el, -end, -ere, -estre, -icge
NounCOM(‘X’)
mid-, samNoun
DIM(‘X’) -incel
Noun
DIST(‘X’) -el, -ful, twiNoun
EFF(‘X’) -bora, -el, -end, -ere, -estre, -ig, -ing, -ling
Noun
ENT(‘X’) -el, -els, -en, -end, -ere, -ett, -ing, -ling, -nes, -ð
Noun
EX(‘X’) and-, be-, ed-, -el, -els, -en, -ere, -ett, for-, -icge, -ig, -ing, -ling,
ō-, of-, on-, -sum, tō-, -ð, ðurh-, under-, -ung, ūp-, wið-, ymbNoun
FEM(‘X’) -el, -en, -esse, -estre, -icge, -ð
Noun
I(‘X’) ā-, ǣ-, æt-, and-, be-, ed-, -el, -els, -en, -end, -estre, -ett, for-,
fore-, forð-, fram-, -hād, -icge, in-, -ing, -ling, -nes, ō-, of-, ofer-,
on-, -rǣden, -scipe, tō-, -ð, under-, -ung, ūp-, ūt-, -wist, ymbNoun
LIKE(LOC(‘X’))
æfter-, æt-, and-, be-, el-, fore-, forð-, fram-, gēan-, in-, of-,
ofer-, on-, tō-, ūp-, ūt-, wiðerNoun
LIKE(TEMP(‘X’))
æfter-, ed-, foreNoun
LOC(‘X’) æfter-, and-, be-, for-, fore-, forð-, in-, mid-,of-, ofer-, on-, under-,
ūp-, ūt-, ymbNoun
MAGN(‘X’) arce-, fore-, forð-, fram-, frēa-, of-, ofer-, sinNoun
MASS(‘X’) and-, -en, -estre, -icge, -iht, -ð, -ung
Noun
MIN(‘X’) -incel, sub-, underNounOPP(‘X’) ante-, unNoun
PAT(‘X’) -el, -end, -ere, -estre, -ing, -ling, wiðerNoun
PEJ(‘X’) for-, misNoun
PRED(‘X’) -dōm, -el, -els, -en, -end, -ere, -ett, forð-, -hād, -ing, -lāc, -ling, -nes,
on-, -rǣden, -scipe, -sum, ð, -ung, -wist
Noun
PRIV(‘X’)
ā-, ǣ-, of-, or-, wanNoun
PROP(‘X’) -dōm, -el, -els, -en, -end, -ere, -ett, -hād, -ing, -ling, -nes, -rǣden,
-scipe, -ð, -ung
Noun
TEMP(‘X’) æfter-, ed-, fore-, mid-, ofer-, sin
Figure 2: Lexical functions by affix and class (non-recursive formations).
RLyLA Vol. 09 (2014), 106-116 | 112
Raquel Vea Escarza
Figure 2 evinces two types of lexical functions which can be distinguished on the grounds of the affixation
process that realizes them: split and unified. Split functions can be realized by both prefixes and suffixes whereas
unified functions opt for prefixation or suffixation. The only pair of semantically complementary functions that
clearly resists this classification is MIN(‘X’), which is realized prefixally and suffixally, and MAGN(‘X’), which is
always applied in prefixation. This is shown in example (6):
(6)MAGN(‘BISCEOP’): arcebiscop ‘archbishop’
MIN(‘DĪACON’): subdīacon ‘subdeacon’
It is usually the case that a pair of semantically complementary functions is split in the formation of a category
and unified in the formation of the other. This happens to LOC(‘X’) and LIKE(LOC(‘X’)), which are split in adjective
formation but represent unified functions in noun formation. Example (7a) illustrates this question for nouns, and
example (7b) for adjectives:
(7)a.LIKE(LOC(‘RYNE’)): æfterryne ‘an encountering’
LOC(‘LYFT’): ūplyft ‘upper air’
b.LOC(‘DÆL’): ofdæl ‘inclined (downwards)’
LIKE(LOC(‘TŌ’)): tōweard 1 ‘facing’
This said, several generalizations might be made. The split functions that can consistently be expressed prefixally
and suffixally both in adjective and noun formation are the function I(‘X’), which relates two partial synonyms to
each other, the associative functions EX(‘X’), WITHENT(‘X’) and WITHPROP(‘X’), and the quantification functions
DIST(‘X’) and MASS(‘X’), which can be realized by prefixes and suffixes, but it occurs in noun formation only. The
examples in (8a) include nouns, whereas those in (8b) include adjectives:
(8)a.I(‘HERE’): herescipe ‘troop’
EX(‘(GE)NEFA’): fōrnefe ‘nephew’s daughter’
DIST(‘HAND’): handful ‘handful’
MASS(‘STRĒAW’): gestrēagung ‘vegetation’
b.
I(‘HOLD 1’): holdlic ‘faithful’
EX(‘FELL’): fellen ‘made of skins’
WITHENT(‘BLŌSTM’): blōstmbǣre ‘flower-bearing’
WITHPROP(‘FǢTT 2’): fǣted 1 ‘ornamented with gold’
DIST(‘GĒAR’): ðrigēare ‘three years old’
As for unified functions, in adjective formation, the following lexical functions are realized by prefixes only:
COUNTFACT(‘X’) (un-), INTENS(‘X’) (eall-, for-, forð-, frēa-, ful-, ofer-, on-, sin-, ðurh-, ūt-), MIT(‘X’) (healf-, mid-,
sām-), OPP(‘X’) (am-, and-, in-, un-), PEJ(‘X’) (mis-) and TEMP(‘X’) (æfter-, for-, fore-). Illustrations of these functions
appear under (9):
(9)COUNTFACT(‘(GE)SELLAN’): unseald ‘ungiven’
INTENS(‘HĀLIG 1’): eallhālig ‘all-holy’
MIT(‘GENACOD 2’): healfnacod ‘half-naked’
OPP(‘SEALT 2’): unsealt ‘unsalted’
PEJ(‘MICEL 1’): mismicel ‘of varying sizes’
TEMP(‘(GE)BERAN’): æfterboren ‘afterborn’
In noun formation, the functions that are realized by prefixes only include COM(‘X’) (mid-, sam-), LIKE(LOC(‘X’))
(æfter-, æt-, and-, be-, el-, fore-, forð-, fram-, gēan-, in-, of-, ofer-, on-, tō-, ūp-, ūt-, wiðer-), LIKE(TEMP(‘X’))
(æfter-, ed-, fore-), LOC(‘X’) (æfter-, and-, be-, for-, fore-, forð-, in-, mid-, of-, ofer-, on-, under-, ūp-, ūt-, ymb-),
MAGN(‘X’) (arce-, for-, fore-, forð-, fram-, frēa-, of-, ofer-, sin-), OPP(‘X’) (ante-, un-), PEJ(‘X’) (for-, mis-), PRIV(‘X’)
(ǣ-, of-, or-, wan-), TEMP(‘X’) (æfter-, ed-, fore-, mid-, ofer-, sin-). This can be seen in (10):
(10)COM(‘HELP’): midhelp ‘assistance’
LIKE(LOC(‘HERE’)): innhere ‘native army’
LIKE(TEMP(‘WORD 1’)): foreword ‘condition’
LOC(‘LǢS 2’): ūtlǣs ‘out-pastures’
MAGN(‘MANN’): forðman ‘man of rank’
OPP(‘LAND’): unland ‘desert’
PEJ(‘LĀR’): mislār ‘ill teaching’
PRIV(‘SPĒD’): wanspēd ‘poverty’
TEMP(‘GIELD’): æftergyld ‘further payment’
RLyLA Vol. 09 (2014), 106-116 | 113
Raquel Vea Escarza
To continue with unified functions, in adjective formation, the following lexical functions are realized by suffixes
only: LIKE(‘X’) (-ed, -en, -ig, -iht, -lic), REL(‘X’) (-ig, -lic) and STA(‘X’) (-el, -end, -ende, -ing, -ol, -or). These functions
are illustrated in (11).
(11)LIKE(‘ÆPPEL’): æppled ‘shaped like an apple’
REL(‘CEOSEL’): ceoslig ‘gravelly’
STA(‘SLĪPAN’): slipor ‘slippery’
In noun formation, the following lexical functions result from suffixation only: AG(‘X’) (-el, -end, -ere, -estre,
-icge), DIM(‘X’) (-incel), EFF(‘X’) (-bora, -el, -end, -ere, -estre, -ig, -ing, -ling), ENT(‘X’) (-el, -els, -en, -end, -ere, -ett,
-ing, -ling, -nes, -ð), FEM(‘X’) (-el, -en, -esse, -estre, -icge, -ð), PROP(‘X’) (-dōm, -el, -els, -en, -end, -ere, -ett, -hād,
-ing, -ling, -nes, -rǣden, -scipe, -sum, -ð, -ung). Example (12) provides an illustration of each of these functions:
(12)AG(‘(GE)ĒHTAN’): ēhtere ‘persecutor’
DIM(‘SCIP’): scipincel ‘little ship’
EFF(‘SWEORD 1’): sweordbora ‘sword-bearer’
ENT(‘HRIMPAN’): hrympel ‘wrinkle’
FEM(‘GOD 1’): gyden ‘goddess’
PROP(‘HĀLIG 1’): hāligdōm ‘holiness’
It follows from this description of unified functions that PEJ(‘X’) and TEMP(‘X’) are realized exclusively through
prefixation, both in adjective and noun formation. The phenomenon of negation, associated with the lexical
functions COUNTFACT(‘X’) and OPP(‘X’), is clearly prefixal. As for the negative function PRIV(‘X’), it is exclusively
prefixal in noun formation and prefixal with the only exception of the suffix -lēas in adjective formation. As it
turns out, in adjective formation the expression of circumstances of time, place, and company remains mainly
prefixal and the type of modification conveyed by intensification and mitigation is also a product of prefixation.
Pairs of a non-figurative and a figurative function, including LOC(‘X’) and LIKE(LOC(‘X’)) as well as TEMP(‘X’) and
LIKE(TEMP(‘X’)) are also realized by means of prefixation, the former in noun formation only. Other logically related
pairs like INTENS(‘X’) and MIT(‘X’) are found in prefixation only.
If we look at suffixation, deverbal nominalizations based on the first argument, including AG(‘X’), EFF(‘X’),
FEM(‘X’), and deverbal nominalizations based on the second argument or non-arguments, including STA(‘X’),
ENT(‘X’) and PROP(‘X’), consistently opt for suffixal means. PRED(‘X’), which belongs in this group, is consistently
suffixal too, leaving aside one very exceptional instance of prefixation with on-. Furthermore, denominal
adjectivizations, as is the case when the lexical function LIKE(‘X’) applies, are restricted to suffixation.
5. CONCLUSIONS
This article has analyzed the prefixation and suffixation of Old English nouns and adjectives on the grounds of
an extended framework of lexical functions that has been enlarged and adapted to the empirical data by drawing
on the structural-functional theories of language. The following conclusions can be drawn.
Firstly, there are more functions that rely on prefixation exclusively than there are functions restricted to
suffixation, which means that a wider array of meanings can be expressed by prefixation and, more importantly,
that the meaning conveyed, as a general rule, by the instances of prefixation under analysis is more contentful,
as opposed to the meaning contributed by suffixation, which, with some exceptions, is more structural. In other
words, prefixation tends to add new senses or aspects of meaning, whereas suffixation tends to change lexical
class, with which the general characteristics of the output class (stating, for instance, that adjectives express
states or qualities) are an essential component of the meaning of the derivative.
Secondly, prefixation is meaning oriented while suffixation is class oriented. By meaning oriented it is meant
that the main target of the derivational process is the modification of meaning, in such a way that the meaning
of the derivative is less predictable from the general characteristics of the output category. By class oriented it is
meant that the main target of the derivational process is the modification of lexical class, so that the meaning of
the derivative is more predictable from the general characteristics of the output category.
REFERENCES
Beard, R. (1995). Lexeme-Morpheme Base Morphology. Albany: State University of New York Press.
Beard, R. and Volpe, M. (2005). “Lexeme-Morpheme Base Morphology”, in P. Stekauer and R. Lieber (eds.),
Handbook of Word-Formation. Dordrecht: Springer. pp. 189-205. doi:10.1007/1-4020-3596-9_8
RLyLA Vol. 09 (2014), 106-116 | 114
Raquel Vea Escarza
Bosworth, J. and Toller, T. N. (1973). An Anglo-Saxon Dictionary. Oxford: Oxford University Press.
Butler, C. (2003a). Structure and Function: A Guide to the Three Major Structural-Functional Theories, Part 1:
Approaches to the simplex clause. Amsterdam: John Benjamins.
Butler, C. (2003b). Structure and Function: A Guide to the Three Major Structural-Functional Theories, Part 2: From
clause to discourse and beyond. Amsterdam: John Benjamins.
Clark Hall, J. R. (1996). A Concise Anglo-Saxon Dictionary. Toronto: University of Toronto Press.
Dik, S. (1997a). The Theory of Functional Grammar. Part 1: The Structure of the Clause. Edited by K. Hengeveld.
Berlin: Mouton de Gruyter.
Dik, S. (1997b). The Theory of Functional Grammar, Vol. 2: Complex and Derived Constructions. Edited by K.
Hengeveld. Berlin: Mouton de Gruyter.
Foley, W. and Van Valin, R. (1984). Functional Syntax and Universal Grammar. Cambridge: Cambridge University
Press.
Goddard, C. and Wierzbicka, A. (2002). “Semantic Primes and Universal Grammar”, in C. Goddard and A.
Wierzbicka (eds.), Meaning and Universal Grammar. Theory and Empirical Findings (Volume I). Amsterdam:
John Benjamins. pp. 41-85. doi:10.1075/slcs.60.08god
Guarddon Anelo, M. C. (2009a). “The Natural Semantic Metalanguage of Old English Compound Adpositions”, ES.
Revista de Filología Inglesa, 30: 61-84.
Guarddon Anelo, M. C. (2009b). “Un análisis de las propiedades combinatorias de los primitivos semánticos a
través de las adposiciones complejas en inglés antiguo”, Revista Española de Lingüística, 39/2: 93-122.
Haselow, A. (2011). Typological Changes in the Lexicon. Analytic Tendencies in English Noun Formation. Berlin:
Mouton de Gruyter. doi:10.1515/9783110238211
Healey, A. diPaolo, ed. (2008). The Dictionary of Old English in Electronic Form A-G. Toronto: Dictionary of Old
English Project, Centre for Medieval Studies, University of Toronto.
Healey, A. diPaolo, (ed.), with J. Price Wilkin and X. Xiang. (2004). The Dictionary of Old English Web Corpus.
Toronto: Dictionary of Old English Project, Centre for Medieval Studies, University of Toronto.
Heidermanns, F. (1993). Etymologisches Wörterbuch der germanischen Primäradjektive. Berlin: Walter de Gruyter.
doi:10.1515/9783110871616
Hengeveld, K. and Lachlan Mackenzie, J. (2008). Functional Discourse Grammar. A typologically-based theory
of language structure. Oxford: Oxford University Press. doi:10.1093/acprof:oso/9780199278107.001.0001
Holthausen, F. (1963). Altenglisches etymologysches Wörterbuch. Heidelberg: Carl Winter Universitätsverlag.
Kastovsky, D. (1992). “Semantics and vocabulary”, in R. M. Hogg (ed.), The Cambridge History of the English
Language I: The Beginnings to 1066. Cambridge: Cambridge University Press. pp. 290-408. doi:10.1017/
CHOL9780521264747.006
Kastovsky, D. (2006). “Typological Changes in Derivational Morphology”, in A. van Kemenade and B. Los (eds.), The
Handbook of the History of English. Oxford: Blackwell publishing. pp. 151-177. doi:10.1002/9780470757048.
ch7
Lacalle Palacios, M. (2011). “The derivational map of Old English and the limits of gradual derivation”, Revista de
Lingüística y Lenguas Aplicadas, 6: 243-259.
Lass, R. (1994). Old English: a Historical Linguistic Companion. Cambridge: Cambridge University Press.
doi:10.1017/CBO9780511621000
Martín Arista, J. (2008). “Unification and separation in a functional theory of morphology”, in R. D. Van Valin, Jr.
(ed.), Investigations of the Syntax-Semantics-Pragmatics Interface. Amsterdam: John Benjamins. pp. 119145. doi:10.1075/slcs.105.12mar
Martín Arista, J. (2009). “A Typology of Morphological Constructions”, in C. Butler and J. Martín Arista (eds.),
Martín Arista, J. (2010a). “OE strong verbs derived from strong verbs”, SKASE Journal of Theoretical Linguistics,
7: 36-56.
Martín Arista, J. (2010b). “Lexical negation in Old English”, NOWELE (North-Western European Language Evolution),
60/61: 89-108.
Martín Arista, J. (2011a). “Morphological relatedness and zero alternation in Old English”, in P. Guerrero Medina
(ed.), Morphosyntactic Alternations in English. Functional and Cognitive Perspectives. Sheffield: Equinox.
pp. 339-362.
Martín Arista, J. (2011b). “Adjective formation and lexical layers in Old English”, English Studies, 92/3: 323-334.
doi:10.1080/0013838X.2011.564776
Martín Arista, J. (2011c). “Projections and constructions in functional morphology: the case of HRĒOW», Language
and Linguistics, 12/2: 393-425.
RLyLA Vol. 09 (2014), 106-116 | 115
Raquel Vea Escarza
Martín Arista, J. (2012a). “The Old English Prefix ge-: A Panchronic Reappraisal”, Australian Journal of Linguistics,
32/4: 411-433. doi:10.1080/07268602.2012.744264
Martín Arista, J. (2012b). “Morfología flexiva en RRG”, in R. Mairal, L. Guerrero and C. González (eds.), El
Martín Arista, J. (2013). “Recursivity, derivational depth and the search for Old English lexical primes”, Studia
Neophilologica, 85/1: 1-21. doi:10.1080/00393274.2013.771829
Martín Arista, J. and Cortés Rodríguez, F. (Forthcoming). “From directionals to telics: meaning construction, wordformation and grammaticalization in Role and Reference Grammar”, in M. A. Gómez González, F. Ruiz de
Mendoza Ibáñez and F. Gonzálvez García (eds.), Form and function in language: Functional, cognitive and
applied perspectives. Amsterdam: John Benjamins.
Martín Arista, J. and Martín de la Rosa, V. (2006). “Old English Semantic Primes: Substantives, Determiners and
Quantifiers”, ATLANTIS, 17: 9-28.
Mateo Mendaza, R. (2013). “The Old English exponent for the semantic prime TOUCH. Descriptive and
methodological questions”, Australian Journal of Linguistics, 33/4: 449-466. doi:10.1080/07268602.2013
.857574
Mel’čuk, I. (1989). “Semantic primitives from the Viewpoint of the Meaning-Text Linguistic Theory”, Quaderni di
semantica, 10/1: 65-102.
Mel’čuk, I. (1996). “Lexical Functions: A Tool for the Description of Lexical Relations in the Lexicon”, in L. Wanner
(ed.), Lexical functions in lexicography and natural language processing. Amsterdam: John Benjamins. pp.
37-102.
Mel’čuk, I. (2006). Aspects of the Theory of Morphology. Edited by D. Beck. Berlin: Mouton de Gruyter.
Mel’čuk, I., Clas, A. and Arbatchewsky-Jumarie, N. (1984–99). Dictionnaire explicatif et combinatoire du français
contemporain: recherches lexico-sémantiques I–IV. Montréal: Presses de l’Université de Montréal.
Orel, V. (2003). A Handbook of Germanic Etymology. Leiden: Brill.
Pokorny, J. (1989). Indogermanisches etymologisches Wörterbuch. Bern: Francke.
Pounder, A. (2000). Processes and Paradigms in Word-Formation Morphology. Berlin: Mouton de Gruyter.
doi:10.1515/9783110814378
Reibel, D. A. (1963). A Grammatical Index to Compound Nouns of Old English Prose. Ph.D. dissertation.
Bloomington: Indiana University.
Seebold, E. (1970). Vergleichendes und etymologisches Wörterbuch der germanischen starken Verben. The Hague:
Mouton. doi:10.1515/9783110821956
Sweet, H. (1976). The Student´s Dictionary of Anglo-Saxon. Cambridge: Cambridge University Press.
Torre Alonso, R. (2009). Morphological Process Feeding in the Formation of Old English Nouns: Zero Derivation,
Affixation and Compounding. Ph.D. Dissertation, Department of Modern Languages, University of La Rioja.
Torre Alonso, R. (2011a). “The Morphological Structure of Old English Complex Nouns”, ATLANTIS, 33/1: 127-146.
Torre Alonso, R. (2011b). “Nominal Affix Combinations in Old English: Distribution and Restrictions”, RESLARevista Española de Linguística Aplicada, 24: 257-279.
Torre Alonso, R. (2012). “Recursive suffixation in the formation of Old English nouns”, in J. Martín, R. Torre Alonso,
A. Canga Alonso and I. Medina Barco (eds.), Convergent Approaches on Mediaeval English Language and
Literature. Newcastle: Cambridge Scholars. pp. 63-86.
Trips, C. (2009). Lexical Semantics and Diachronic Morphology. The development of -hood, -dom and -ship in the
History of English. Tübingen: Max Niemeyer Verlag. doi:10.1515/9783484971318
Van Valin, R. (2005). Exploring the Syntax-Semantics Interface. Cambridge: Cambridge University Press.
doi:10.1017/CBO9780511610578
Van Valin, R. and LaPolla, R. (1997). Syntax: Structure, meaning and function. Cambridge: Cambridge University
Press. doi:10.1017/CBO9781139166799
Wierzbicka, A. (2002). “Semantic Primes and Linguistic Typology”, in C. Goddard and A. Wierzbicka (eds.), Meaning
and Universal Grammar. Theory and Empirical Findings (Volume II). Amsterdam: John Benjamins. pp. 257300. doi:10.1075/slcs.61.10wie
RLyLA Vol. 09 (2014), 106-116 | 116
Vol. 9 año 2014, 117-119
EISSN 1886-6298
Patricio Moya Muñoz
Universidad Politécnica de Valencia
Reseña de Mancera, A. y Pano, A. (2013). El discurso político en Twitter: Análisis de mensajes que
“trinan”. Barcelona: Anthropos
El creciente aumento en el uso de las redes sociales virtuales desarrolladas en las nuevas tecnologías
de la información (como Facebook, Twitter, o Tumblr) ha promovido un tipo de intercambio comunicativo
con características inherentes a este tipo de medios. En otras palabras, las maneras en que los usuarios se
relacionan se encuentran determinadas, entre otras razones, por las constricciones que impone el contexto
multimedia: las nuevas tecnologías facilitan un diálogo constante, en la medida en que existe una proximidad
comunicativa y una sensación de co-presencia que los usuarios manifiestan al momento de vincularse con otros
(Pano, 2008). Lamentablemente, en la actualidad, a pesar de los valiosos esfuerzos por estudiar el Discurso
Mediado por Ordenador (en adelante DMO) en español (Yus, 2010; Pano, 2008, entre otros) falta mucho por hacer.
Específicamente, el estudio de las interacciones que se llevan a cabo en Twitter, se han abordado, principalmente,
con tuits en idioma inglés y desde la perspectiva de la lingüística computacional. Esto ha decantado en que los
esfuerzos se hayan concentrado, en general, en encontrar regularidades estadísticas a partir de corpus de cientos
de miles de tuits, con el objetivo de lograr, por ejemplo, algún tipo de impacto en el área de las encuestas políticas
de opinión, al intentar predecir una elección presidencial (Tumasjan, Sprenger, Sandner y Welpe, 2010) o en
realizar estudios de mercado, empleando Web Opinion Mining (Ghiassi, Skinner y Zimbra, 2013) para especificar
los sentimientos de los usuarios frente a un determinado producto.
En este sentido, el presente libro escrito en conjunto por Ana Mancera, profesora de la Universidad de Sevilla,
y Ana Pano, profesora de la Universidad de Bolonia, viene a profundizar una línea de investigación contingente
y dinámica, que afronta este tipo de intercambios comunicativos desde una perspectiva pragmática y con una
metodología cualitativa que enfatiza la función y efectos sociales del lenguaje producido en las redes sociales
virtuales. En específico, las autoras estudian la forma en que algunos políticos españoles interactúan tanto con
la audiencia como entre sí en Twitter durante la campaña presidencial española del año 2011. Para el análisis, se
recogieron 454 tuits de cuentas de Twitter de: candidatos, importantes periodistas del área política, ciudadanos
en general y, por último, de cuentas que parodiaban a los candidatos presidenciales.
El libro se encuentra dividido en tres partes. En la primera de ellas, se lleva a cabo un breve estado del
arte sobre el DMO. El primer capítulo de esta sección detalla, principalmente, los cambios culturales que ha
implicado esta nueva manera de comunicación en la Web 2.0, en tanto el individuo que emplea estos medios
manifiesta una noción del yo no pasivo, es decir, un sujeto que es protagonista de los hechos y, al mismo tiempo,
genera contenidos sobre los mismos. En cuanto al tipo de lenguaje empleado, éste no tan solo se ajusta a los
fines comunicativos que favorece este tipo de medios (comunicarse y compartir con otros usuarios) sino que los
mecanismos audiovisuales que acompañan a la comunicación se erigen como una ayuda indispensable para
incrementar las relaciones entre los sujetos. La preferencia por el lenguaje coloquial por sobre otros registros
permite, asimismo, lograr mayor empatía tanto con conocidos como con desconocidos, dado que se produce una
dinámica actualización de las actividades de los sujetos, actividades que son comentadas o replicadas por los
demás usuarios; por tanto, las invetigadoras aseveran que la forma de comunicación en la Web 2.0 es el diálogo
virtual entre los participantes. Como consecuencia de lo anterior, los usuarios se han empoderado del mecanismo
de generación de contenido al desplazar el sitio prototípico en el que se originaba la opinión pública: desde un
espacio abierto (salones domésticos y cafés, en palabras de Habermas) a un ambiente virtual en el que todos los
ciudadanos pueden participar, lo que simboliza una apertura de la política tradicional.
El segundo capítulo de la primera parte se centra en los orígenes y características principales de Twitter. Twitter
se abre al público en julio del año 2006 e intenta ser un entorno virtual que facilitara la transmisión de información
en tiempo real. A partir de abril de 2008, su crecimiento es exponencial: los 400.000 tuits por trimestre que se
| 117
Reseña de Mancera, A. y Pano, A. (2013). El discurso político en Twitter: Análisis de mensajes que “trinan”
escribían en 2007 aumentaron a 200 millones diarios en junio de 2010, gracias a su capacidad de comunicar
sucesos periodísticos de manera viral. La principal característica de Twitter radica en su capacidad de fomentar
un tipo de comunicación que valora la brevedad de los mensajes, reflejando el antiguo concepto de economía
lingüística: intentar dar cuenta de la mayor cantidad de información con el menor uso de recursos. Para lograrlo,
los usuarios se valen tanto una serie de abreviaturas que cumplen una función determinada dentro del intercambio
comunicativo (RT, DM, @, TT, CC) como etiquetas antepuestas con el signo # (denominados hashtags). Las
autoras afirman que en Twitter “el poder de la conversación en tal contexto no está tanto en la profundidad de los
temas tratados y de los mensajes compartidos, sino en su poder viral, en su capacidad para canalizar y difundir la
información a niveles nunca antes conseguidos” (p. 50). Por tanto, es esta característica la que convierte a Twitter
en una herramienta política tan potente, en la medida en que facilita la participación pública en la vida política:
promueve, de esta manera y teóricamente, la conversación directa y en tiempo real entre políticos y ciudadanos.
En la segunda parte, las autoras trazan las líneas que articulan su investigación. El capítulo que abre esta
sección da cuenta de las características político-sociales que se vivían en España en el momento de la campaña
y posterior elección presidencial del año 2011. La alta cifra de parados, la cual según Zapatero se reduciría, y la
corrupción se convirtieron en los hechos que marcaron el desarrollo de la elección. Las redes sociales también
hicieron su parte: gracias a Facebook o Twitter, se impulsó la coordinación entre las personas que se manifestaron
en distintos lugares del país a partir del 15 de mayo de 2011. Como sea, a juicio de las investigadoras, el papel
que cumple la televisión es fundamental para los políticos (más que los periódicos o las mismas redes sociales),
en tanto “un porcentaje muy elevado de ciudadanos no recibe más contenidos políticos que los que le llegan por
la televisión” (p.68); de ahí que los debates televisados alcancen una relevancia fundamental para las cadenas de
televisión.
El siguiente capítulo de esta sección examina las similitudes que existirían entre las campañas llevadas a cabo
en Twitter por los políticos españoles en relación con la desarrollada por Barack Obama y su comando durante el
año 2008, considerada como la mejor campaña desde el punto de vista mediático, en la medida en que integró a
los tanto a los ciudadanos como a sus prerrogativas. La campaña del equipo de Obama fue persona a persona,
gracias al empleo de tecnologías de comunicación que en ese momento eran emergentes como Youtube, Facebook
o Twitter. En el caso de esta última, el grado de compromiso con los ciudadanos que logró la cuenta de Obama
es considerada, al menos, notable, dado que seguía casi al mismo número de personas que leía sus mensajes. Lo
anterior, junto con que cada mensaje enviado por algún usuario era respondido, produjo una sensación de gran
proximidad con el candidato. A pesar de lo anterior, las características del electorado estadounidense es muy
diferente al español, por lo que independiente de los esfuerzos realizados por los partidos políticos españoles por
imitar el modelo de campaña de Obama, “la cifra de penetración de la tecnología digital entre el electorado es
cuatro veces inferior a la norteamericana” (p.83). De esta manera, los mensajes enviados por Twitter de parte de
los candidatos tenían un fin, casi exclusivamente, propagandístico. Por último, el tercer capítulo, que cierra esta
sección, destaca el papel de Twitter en la publicación y masificación de la opinión pública. Con este fin, las autoras
investigaron 237 tuits enviados por respetados actores del mundo periodístico (connotados periodistas, editores
de periódicos, entre otros) que cubrían la campaña electoral. El uso que éstos daban a la red social era muy
variado, sin embargo, destaca el empleo meramente valorativo de alguna información que les parecía relevante,
al renviar algún enlace acompañado de su opinión.
La tercera parte del libro se encarga estudiar desde un punto de vista pragmalingüístico el corpus recogido para
la investigación. El primer capítulo analiza los mensajes de los políticos españoles, considerando los mecanismos
empleados para dialogar con los ciudadanos por medio de los rasgos inherentes al canal empleado. El análisis
se llevó a cabo teniendo en cuenta los rasgos prototípicos del lenguaje político, es decir: el carácter ambiguo
del lenguaje; el carácter polémico, con el fin de construir un adversario; la intención agitadora, empleada para
convencer al electorado. En cuanto a la ambigüedad, el uso de palabras y expresiones vagas, o semánticamente
equívocas, es muy utilizado por los candidatos, quienes emplean conceptos (como ‘valor’) cuya interpretación
dependerá de la ideología a la que representan. Asimismo, el elevado empleo de eufemismos (desaceleración
económica) y de nominalizaciones de sustantivos deverbales (reducción, recuperación, abaratamiento) implica
una intención por parte de los políticos por no comprometerse con lo dicho. En segundo lugar, el carácter polémico
se ve reflejado en todo discurso por la manera en que los diferentes participantes del mundo textual participan en
cada interacción, en otras palabras, se relaciona con la polifonía textual. De esta manera, mientras la utilización
del yo se encuentra muy presente en la repetición de discursos ya pronunciados anteriormente, el empleo de
vosotros/ustedes responde a condicionantes de índole pragmática: el primero es empleado para referirse o
dirigirse a los simpatizantes mientras que ustedes se reserva, esencialmente, para los adversarios políticos. En
tercer lugar, la intención agitadora se manifiesta de diferentes maneras dentro del corpus analizado por medio
del uso de metáforas (“esa receta es la misma la de Cameron”), expresiones idiomáticas (“seguiremos dando
el callo”), repetición (en este caso, de mensajes completos), entre otras. No obstante, uno de los recursos más
interesantes llevados a cabo por los candidatos es el de la entextualización, esto es, “extraer fuera de contexto
otras voces con el fin de modificar su contenido original y adaptarlo a los intereses del hablante” (p. 171).
RLyLA Vol. 09 (2014), 117-119 | 118
Reseña de Mancera, A. y Pano, A. (2013). El discurso político en Twitter: Análisis de mensajes que “trinan”
El segundo capítulo de la tercera parte investiga el uso creativo (que permite las características del medio) de
la puntuación para establecer relaciones entre los diferentes participantes de la comunicación. La estrategia más
empleada es contestar a los usuarios mencionándolos (utilizando el signo @) en la conversación, lo que da una
impresión de cercanía y preocupación. Además, el retuiteo (RT) de los mensajes de usuarios apoya una función de
reconocimiento, al dar relevancia a lo que se está enviando por parte del candidato. El uso de etiquetas a través
de hashtags (#) promueve que estos mismos se transformen en eslóganes en algún momento de la campaña
(#votaporloquequieres). El examen del uso de estas estrategias por parte de los candidatos indicó a las autoras
que los políticos no aprovechan las ventajas del medio que están utilizando, vale decir, en la mayor parte de los
casos se refieren a sí mismos o a sus contrincantes, en lugar de incluir al resto de los ciudadanos. Por otro lado,
el tercer capítulo estudia las estrategias de cortesía y descortesía presente en los tuits recogidos. Las autoras
aseguran que el estudio de las estrategias empleadas dependerá del rol de cada uno de los participantes en la
interacción. En cuanto a la cortesía, en el corpus estudiado las estrategias empleadas tienden a reforzar la imagen
(cortesía valorizante) que se puede manifestar de manera directa como indirecta. Sin embargo, a pesar de que
existen casos de mensajes que refuerzan la imagen, lo normal en el discurso político es deteriorar la imagen
del otro, por lo que la descortesía es la estrategia predominante. En este sentido, la exhortación fue uno de los
mecanismos más recurrentes (“@marianorajoy Pero diga ya lo q va a hacer coñoooo, que lleva media hora y no
dice nada, menudo presidente siestero”).
El capítulo final busca determinar las estrategias irónicas empleadas por ciertos usuarios, en particular aquellos
que parodiaban a algunos de los candidatos presidenciales (como @Naniano_Rajoy que se burlaba, entre otras
cosas, de la pronunciación del entonces presidenciable del Partido Popular). En cualquier análisis de enunciados
irónicos, lo más difícil resulta determinar los indicadores que establecerían que un enunciado pertenece o no a
dicha categoría. En el corpus revisado, se puede considerar que el uso de emoticonos; signos de puntuación
(como los puntos suspensivos o el signo de interrogación, comillas); manipulación de unidades fraseológicas
(“Pan para Rajoy, hambre para mañana”), entre otros, son indicadores potenciales de ironía. Sin embargo, la
enumeración no es clara y a pesar de todos los indicadores posibles, tal como lo señalan las autoras, sin una
base de conocimiento común que permita identificar las intenciones de los usuarios, es muy difícil establecer si
un enunciado es o no irónico.
En conclusión, la obra presentada se erige como una investigación novedosa que es capaz de dar cuenta de
un tipo de interacción lingüística actual y contingente, a partir de una perspectiva social que permite dar ciertas
luces sobre los tipos de relaciones sociales que se pueden llegar a establecer en Twitter. Es una obra acabada y
documentada. Las referencias bibliográficas empleadas son muy actuales y están en consonancia con lo que se
espera de los estudios en este tipo de contexto: dinamismo constante, dadas las características del medio de
comunicación. Aunque hay aspectos que han quedado abiertos (opinión que las mismas autoras comparten al
final de su libro), hay convencimiento de que la obra reseñada se convertirá, en los años posteriores, en una obra
de referencia obligatoria, gracias tanto la calidad y pertinencia de sus análisis como a la capacidad para llevar a
cabo una investigación con sólidas bases dentro de un contexto en constante cambio.
Referencias
Ghiassi, M., Skinner, J. y Zimbra, D. (2013). “Twitter brand sentiment analysis: A hybrid system using n-gram
analysis and dynamic artificial neural network”, Expert Systems with applications, 40/16: 6266-6282.
doi:10.1016/j.eswa.2013.05.057
Mancera, A., y Pano, A. (2013). El discurso político en Twitter. Análisis de mensajes que “trinan”. Barcelona:
Anthropos.
Pano, A. (2008). Dialogar en La Red. La lengua española en chats, emails, foros y blogs. Berna/Frankfurt: Peter
Lang.
Tumasjan, A., Sprenger, T., Sandner, P. y Welpe, I. (2010). “Predicting Elections with Twitter: What 140 Characters
Reveal about Political Sentiment”. Proceedings of the Fourth International AAAI Conference on Weblogs
and Social Media. pp. 178-185. California.
Yus, F. (2010). Ciberpragmática 2.0. Barcelona: Ariel.
RLyLA Vol. 09 (2014), 117-119 | 119

V olumen 9 - PoliPapers

Transcription

Similar documents

libroALTROper TIPO

Pedro Fernández began his international career when he was

Manteca Evacuation Sector (RD17 Zone)

Forecast - Info

TODAY`S READINGS First Reading — When the wicked turn away

Explorando tamaños: globos aromáticos

St. Jerome Church - St. Jerome Catholic Church

Beerbaum Art International diversity

Know the contents - Miami Virtual School

May 8, 2016 - St. James Catholic Cathedral