The mental lexicon. The individual vocabulary of an adult. The acquisition of the lexicon. The mental lexicon of a bilingual

The mental lexicon]

The word 'lexicon' long time has been associated with lexicography. It was viewed as a large dictionary that contains orthographical representation of an enormous number of words related, first of all, alphabetically. The list of words included much information about their meaning, grammatical characteristics and probably pronunciation that helps, however, to establish other kinds of word-relations.

Lexicologists paid special attention to semantic relations of words in a vocabulary system arid traditionally described them as paradigmatic and syntagmatic, antonymic, and synonymic (sec Chapter 7).

Since knowledge of a language and its vocabulary is stored in our heads, research on the ways this storage is done and exercised was relegated to psychologists.

But modern linguistics is marked by a fusion of theoretical general linguistics and psychology, and the term 'lexicon' becomes more and more associated with the mental lexicon. One of the steps necessary for translating conceptual knowledge into linguistic knowledge is retrieving appropriate lexical units from our mental lexicon.

The mental lexicon is a lexical system representation in our mind.

Linguistics has yet to provide a single undisputed working model of the mental lexicon due to its complexity. Now linguistics can only offer different suggestions concerning its

most abstract aspects, for example, the place of the mental lexicon in the general model of language capacity, structure oflexicon, character and number of items included and types of information about them.

ЛИ scholars agree that our minds should contain the same types of information about the word: phonological, orthographical, morphological, semantic and syntactic, and all of them are somehow linked. Otherwise the word would not be understood, retrieved or properly used.

Within the general model of language capacity the mental lexicon should be directly related first of all to lexicalizcd conceptual structures, because words without meaning make no sense. It should also be connected with syntactic and phonological structures.

In order to perform its generative character without which there is no acquisition and growth of the vocabulary, the mental lexicon should also be connected with rules governing correct formation of conceptual, syntactic and phonological structures.

Schematically it may be represented in the following pattern worked out by Jackendoff/Jackendoff 1997:39/:

As it is presented in the scheme, language has three basic components, with the lexicon being attached to all of them.

Due to the lexicon all these components of the mental grammar match. Interruptions in mapping between the components and/or lexicon create problems in language comprehension or use. 1'he lexicon also has information about specific restrictions the word may have. The mental lexicon thus happens to be a deposit of all knowledge about the meaning, grammar and phonology of the word. Thus, boundaries among lexicon, grammar and phonetics are quite artificial, created just for the sake of analytical convenience.

The crucial questions are what is presented in the mental lexicon and how it is structured to provide reliable storage and retrieving from the memory.

It should also be mentioned that traditional linguists have created a successful science by ignoring the numerous interactions between linguistic knowledge and world knowledge as well as the psychological and neurological structures providing it. They have also developed a methodology that works well for phonology, syntax, morphology and certain areas of semantics. But understanding and describing new, psychological areas of lexical semantics requires special methods of investigation that are still in the process of developing. Modern psycholinguistics makes a wide use of psychological methods, like free association tests, and time measuring tests on word recognition as well as traditional linguistic analyses.

It has been established that words in the mental lexicon are usually kept without inflections. Inflectional suffixes are added to stems later, in speech. But derivational affixes are in the mental lexicon, at least as parts of derived words that are stored there.

Nevertheless, scholars still debate the number and character of units stored in the mental lexicon.

One theory argues that only simple words and their multiple properties are stored in our mental lexicon. Among these properties arc: how a word is pronounced, what part of speech it belongs to, what other words it is related to, and how it is spelled. These properties make up separate entries in our mental lexicon, and each of them makes up a separate interface and has a different access. That is with access to a word's acoustic property we are able to find a rhyming word for it or to list some other words with similar sound structures just by using the lexicon's phonetic interface. With access to the part-of-speech meaning of a word we may retrieve thousands of words with the same lexical-grammatical meaning from our memory. Tapping into semantic interface of a word, we may activate and retrieve lots of words semantically related to it. So, in our mind there are multiple vocabularies each of them with different units as their nearest neighbours.

According to this theory, derived lexical units and rules of word-formation are outside the mental lexicon. The mental lexicon is the place for just simple, non-derived words. Derivatives may be somewhere else, for example, they may be part of grammar or of some other component of the language faculty.

But it is also known, and it was mentioned in the previous chapters, that all derived and!

compound words as well as phraseological units have a special idiomatic component that <

can not be deduced from the formal structure of a lexical unit. This fact provides the!

grounds for believing that they should also be memorized and listed in the mental lexicon. \

Moreover, the rules of word-formation listed in morphology are too general to be adequately applied to a concrete word to form an accepted derivative. It makes more sense to enlist the rules of word-formations with all their exceptions and idiosyncrasies in

the mental lexicon. That will add to the model of the mental lexicon its active generative character that we observe when we produce and interpret new words.

Not all derived and compound words and word combinations should be listed in the mental lexicon but only those that cannot be decomposed without changing the meaning of a lexical unit.

Thus, in the mental lexicon alongside simple words there may be some derived and compound ones and even sentences and some texts. There should also be some rules on how these complex units may be decomposed into simple ones or how a great number of well-formed derived words and even phrases with all their idiosyncratic properties can be easily produced or reproduced in speech.

Thousands of words, morphemes and phraseological units as well as rules of their formation should be stored in our mind in some order, otherwise a momentary successful retrieval and recognition would be impossible. The question, however, is how?

There arc many reasons to believe that there are radical differences in quantity, character and organization between words stored in alphabetically organized dictionaries and the words stored in our minds.

No person knows and uses all the words that a large dictionary may contain (see 'The individual vocabulary of an adult' below). Vice versa, each person has much more information about each word that any dictionary may contain. The information about meaning of the word presented in a dictionary is scarce, dry and meager in comparison with the conceptual information. For example, we know which word stands for prototypical item and which for peripheral (cf: sparrow, penguin, ostrich are all birds, but only sparrow is the most typical of them). We may recognize different pronunciations of a word produced by different speakers while a dictionary may give only one variant. Information about combinability of a word is undcrrepresented in any dictionary: a native speaker knows much more information about lexical and grammatical restrictions on word usage which is quite scarce in a dictionary.

Linguists and psychologists collected much data about storing lexical items and rules in

our mind. Retrieval of words from memoiy and checking the activation zones in our mind

by modern equipment give a lot of information about the structure of the mental lexicon.

Thus, it has been proven that different groups of words are stored differently and are

placed in different cortex /ones. That is why some fields for some reason may be

damaged without involving the others. After strokes people may remember the names of

such concepts as 'sphinx' and 'abacus' but not remember the names of fruit and

vegetables /Aitchison 1994:84/. Verbs and nouns, functional and notional words are

stored separately in the mental lexicon, too.

Slips of the tongue are also an important source of this information. For example, since we do not have slips of the tongue for the words that follow each other in a dictionary, like

decrease and decree, there are grounds to believe that in speech production the phonetic interface is not as close to conceptual structures as, for example, the semantic one, where the relatedncss of such words as forks and knives, and shirts and skirts cause quite frequent slips of the tongue.

So, the mental lexicon may be viewed as a structure with a number of distinct modules for different types of information. There are separate modules for syntactic, phonological, morphological and semantic presentations; content words are supposed to be kept separately from functional words, verbs to be kept separately from nouns and derivational affixes separately from inflectional ones.

Yet, the mental lexicon is not only a complex structure of information but it is also a complex system where all these types of information arc somehow connected.

The degree of connection between different interfaces and between lexical units in the semantic semantic interface of the mental lexicon is different: some links are particularly strong, like conneetionss between co-ordinates and collocational links; some links are somewhat weaker, like the connections between some of hyponyms and hyperonyms.

Nevertheless, hierarchical relations are the most important types of word relations for the assembling the words into a structured whole. One theory assumes that a hyponym inherits the properties of its superordinates. To understand and remember a hyponym we do not need to mcmori/c all the features characteristic of a hypcronym, we need to remember only the distinguishing features of hyponyms. So, the inheritance system saves memory space.

Numerous studies of hierarchical taxonomies of words proved that on the folk-level they typically have no more than five levels (see /Cruise 1991:145/) and frequently have fewer. These levels are commonly labelled as follows:

The most significant level of a taxonomy is called generic. This is the level of names of common things and creatures: rose, cat, oak, apple, car, cup. It is the most numerous level, and it is the level whose units are learned first. It is the level, the units of which are predominantly simple, native and the most frequently used names make up prototypical members of the category.

There are also connections between words of different lexical-semantic fields (interfield relations). Some of them, usually referred to as entailment, or presupposition are strong. Here are some examples of this type of semantic relations between groups of different lexical-semantic fields. 'Killing' entails 'dying', if there is a 'killing event', then there is also a 'dying event'. Or, if John is selling his piano it means that John owns a piano. 'Sight presupposes eye, education presupposes learning, journalist presupposes press.