Components Of Word Knowledge Adapted From Nations English Language Essay

Explain and exemplify what is involved in ‘knowing a word.’ What are some of the ways in which a teacher can help learners acquire the appropriate knowledge.

Introduction

ELT has seen a gradual shift from a focus on grammar to lexis. With this there has been an increased interest in vocabulary learning and teaching strategies. One question that has generated interest is, what criteria need to be fulfilled before a word can be considered learnt? In search of a quantifiable definition, various frameworks have been proposed. Amongst the first was by Richards (1976), who suggested seven aspects of word knowledge: syntactic behaviour, semantic value, different meanings, underlying form, derivations, associations and limitations on use. Nation (1990) developed this idea further to include more components and made a distinction between receptive and productive knowledge (see Table 1).

Table 1: Components of word knowledge (Adapted from Nation, 1990: 31)

Meaning

Conceptual (meaning)

Receptive

Productive

What does the word mean?

What word should be used to express this meaning?

Association

Receptive

Productive

What other words does this word make us think of?

What other words could we use?

Form

Spoken form

Receptive

Productive

What does the word sound like?

How is the word pronounced?

Written form

Receptive

Productive

What does the word look like?

How is the word written and spelled?

Use

Frequency

Receptive

Productive

How common is the word?

How often should the word be used?

Appropriateness (Register)

Receptive

Productive

Where would we expect to find this word?

Where can it be used?

Position

Grammar

Receptive

Productive

What part of speech?

What are the parts of the word?

In what patterns does the word occur?

In what patterns must we use the word?

Collocation

Receptive

Productive

What words and types of words can we express before and after the word?

What words or types of words must we use with this word?

Further dimensions have been added to these ideas, most notably by Haastrup and Henriksen (2000), and Qian (2002).

For the purposes of this essay I will focus on the framework presented by Nation (1990). I will look at what is involved in each sub-category of ‘knowing a word’, then give a justification of its inclusion in Table 1, and finally look at some ways that each component can be supported in the classroom.

A word’s meaning

We expect a word to have an exclusive one to one relationship with its meaning. For example, the proper noun, Mount Everest refers to one particular mountain. However with most words the relationship is not so simple. The word mountain, for example, cannot describe all mountains. Here, the word is a label we attach to our concept of what a mountain is like, based on our personal experiences. Hence, it would be more precise to define meaning as a relationship between a word and its concept.

Therefore, by describing the attributes of a concept, we can formulise the meaning of a word. But, as seen with mountain example, this is not an ideal approach, as a majority of words do not have a one to one relationship with their referent. Two mountains may have very different attributes, so the word mountain can have flexible meaning. Most words to a certain extent have this flexibility, something which Aitchison (1994) calls “fuzzy meaning.” The way people deal with this “fuzziness” is explained by the prototype theory, where the most typical example of a concept category is selected and compared against prospective members. The match does not have to be perfect, which explains why words can sometimes be used with slightly different meanings Aithchison (1994).

The prototype theory explains how people relate words to concept categories but not how they deal with fuzziness in meaning between words. A words meaning can only really be understood by comparing it to the meanings of related words (Gairns and Redman 1986). The study of these meaning relationships, called sense relations falls under the scrutiny of Semantics and their more technical terms are defined in Table 2.

Table 2. Sense relations

Synonymy

The relationship between words having a similar meaning, e.g. big, large.

Antonymy

The relationship between words having opposite meaning, e.g. easy, difficult.

Hyponymy

The relationship between words belonging to the same category (superordinates being the more general category and subordinates the more specific).

Meronymy

The relationship between a part and whole, e.g. trigger, gun.

Teaching meaning is clearly challenging and requires a lot of input. When teaching in a context where learners have no exposure to English outside the classroom it can be difficult to maintain motivation. One way that I found effective, was to use oral or written translation (from learners L1 to L2) with authentic contextualized material. Students would discuss and improve their translations to build vocabulary lists with the relationships highlighted. Using translation in a language classroom is generally regarded as anti-modern but it can have several advantages, including helping learners extend their vocabularies into new areas (Heltai 1989).

A word’s associations

Words are related to each other in various ways, sense relations which we looked at earlier are a good example of this. Our mental lexicon uses these relationships to store and organise words in clusters related by concepts. Lexical units are linked by a complex network, the complexity being proportional to the depth of vocabulary knowledge (Aitchison, 1994). The three most important categories of associations are given in Table 3.

Table 3: Word associations.

Type of association

Description

Examples

Clang

Words are related in sound rather than meaning.

brain – train

stamp – clamp

Syntagmatic

Words that occur together sequentially.

commit – crime

gaping – void

Paradigmatic

Words that are semantically related.

stop – halt

hard – soft

Several studies such as Ervin (1961) have shown that, as a L1 speaker matures there is a shift from clang to syntagmatic and then to paradigmatic associations. Although most of this work has focused on native speakers, work done by Meara (1983) and subsequently by Greidanus et al (2001), has found that as L2 proficiency improves, association responses by L2 learners also increase incrementally to resemble those of native speakers.

Although there is still no clear idea what processes are involved in the development of a person’s mental lexicon organisation, what is clear is that syntagmatic associations are common and play an important role in helping a learner move up to more paradigmatic ones. Hence incorporating syntagmatic associations (commonly known as collocations) into vocabulary teaching can be an effective way of strengthening word knowledge and making learners aware of the interconnection between words.

A good way to practice paradigmatic features would be through word classification activities. Students classify a set of target vocabulary into categories using sense relations, they then justify, explain and discuss their decisions. This is probably more relevant for advanced students and it can help to visualise and stimulate the organisation of vocabulary in the mental lexicon.

A word’s spoken form

Knowledge of word meaning is not very useful unless a word can be recognized and produced. The non-meaning aspects of word knowledge, the written, spoken and grammatical forms are equally important. Knowing a word in its spoken form means: being able to distinguish and understand a word in continuous speech, and to produce the word and be understood. Listening, the receptive side of this process involves extracting the word (phonologically) from continuous speech and then using this phonological representation to map it to related lexical knowledge (Schmitt, 2000). While listening, the sounds of phonemes within the word need to be recognised, individually and in combination. Equally essential, is being aware of how the word is split into syllables and recognizing the associated stress patterns. It is this last aspect which makes it difficult for learners of English to segment individual words and separate them within continuous speech. An added difficulty is the fact that learners have little control over the rate at which words are received.

English is a stress-timed language where stressed syllables contain full vowels, and weak syllables have reduced vowels. Knowledge of this distinction could help learners in segmentation. Cutler and Norris (1988) proposed that English listeners expect words to begin with strong syllables and separate continuous speech with maximum effect at the point a strong syllable is encountered. In an experiment they found listeners took longer to distinguish the word mint when embedded in mintayf (strong second vowel), but were quicker to detect it in mintef (reduced second vowel). This idea can be useful in identifying lexical rather than grammatical words as the latter contain weak monosyllables.

Once the word has been extracted as speech segments, the mental lexicon is activated and retrieves the relevant lexical information. One of the ways that the brain deals with this process was proposed by Marslen-Wilson (1987) in the form of the Cohort model. According to this model the lexical retrieval starts as soon as the first speech segment reaches the ear, and all words beginning with this segment are activated in the mind. The process continues with each input segment, mismatches are deleted until only the perfectly matched word remains activated. Context and coarticulation are also taken into account during lexical retrieval (Packard, 2000), and word knowledge depth will influence the speed and accuracy of word recognition (Cole and Jakimik, 1978).

The productive aspects of the spoken form are concerned with the ability to pronounce and use the stress patterns of a word. For the spoken form of a word to be produced successfully it needs to be retrieved accurately from memory, this phonological short-term memory has a major influence on overall vocabulary learning ability (Gathercole and Baddeley, 1989). Differences in L1 and L2 sound systems can negatively affect the efficiency of an individual’s memory. This effect can be controlled by employing meaning based association techniques, like the ‘keyword’ technique to create a phonological and meaning connection between the L2 target word and a known L1 word.

When learning the spoken form, becoming familiar with the word and sentence level stress patterns is of primary importance. At the word level, the teacher can pronounce the word in isolation and then use phonemics and stress markers to present the word visually. However in natural communication many of the phonological characteristics are determined by the context and surrounding speech, so it would be useful if the word is also presented in a wider context.

A word’s written form

With regards to written/orthographical knowledge, there is growing evidence from research proving it plays an important role in vocabulary acquisition (Schmitt, 2000). The receptive component of orthographic knowledge is word recognition: the visual identification of a word by its outline-shape, and an analysis of the letters that it contains. Scientific evidence suggests that the main input into the word recognition process comes from recognising a word’s individual components (Besner and Johnston, 1989). The positioning of the letters within a word is particularly important. In the results of an experiment by Rayner and Hagelberg (1975) it was observed that kindergarten students focused on the first letter to recognise a word, whereas adults used the first and second letters plus the shape of the word.

Further evidence to support the importance of letter positioning is provided by eye movement literature. The eyes do not move smoothly while reading, but fixate on a word for short periods (200-250 milliseconds) before moving on to another word. These movements, called saccades, normally last 20-35ms. Fixations are usually in the first half of a word and not all words are fixated. Short words, especially function words are often ignored. The centre point of a person’s vision, the fovea, can only see 3-4 letters to the left of and about 15 letters past the fixation point. Word recognition takes place closest to the fixation point and information further out is used to identify the length of the following word and the next fixation point. This nature of eye movement suggests that the beginning of a word is the most important in the visual recognition process. As proficiency improves this recognition process becomes faster (Meara, 1986).

The productive side of writing is spelling. Studies have shown that there are 2 stages of spelling development amongst children, phonological processing and orthographic. At the early stages of literacy, young learners depend primarily on phonological processing, and with maturity move towards the orthographic stage, here spelling depends more on the activation of lexical knowledge (Frith, 1985). Other research however, suggests that both these stages occur contiguously from the onset (Lennox and Seigel, 1994).

If, as has been suggested that improved word recognition positively affects language proficiency, then it is a good idea to encourage automaticity of word recognition in the classroom. The following could be an example of a recognition activity (adapted from Morrow, 1980):

Timed activity: Some of these are not really English words. Don’t worry about their meaning. Find the words with the same spelling as the ones in darker print, and underline them.

Immediately

imediately immediatly immediately immediately

experience

experienced expereince exporience experience

automatic

automatic autematic atomatic automate

Word recognition occurs dynamically, using computers can help make this learning more stimulating. For example, the above activity could be made more interactive by having words appearing on screen for increasingly shorter periods to help increase the speed of recognition. Tozcu and Coady (2004) looked at how interactive computer based texts affected vocabulary acquisition, their results showed a positive effect of using technology on vocabulary development.

English spelling irregularity creates difficulty for learners, so one of the first things that need attention is the sound-spelling correspondences. Some useful spelling strategies are given in Table 4.

Table 4: Spelling Strategies

Mental images

Can help with spelling words with phonological- visual disparities, such as those with redundant letters (Upward, 1988). A mental image of a similar known word can also be used to check the spelling of an unfamiliar word (Schmitt, 2000).

Analogy

Using common rimes (e.g. ity, ight ent). For example, by recalling current knowledge of the words: plan, acid and quantity, students can attempt to spell placidity (Nation, 2008).

Word families

Orthographical similarities can be utilised. For example, when learning protect, the derivative forms protected, protecting, protective, protection and protector can also be highlighted. Polysemous (e.g. bank/bank) and compound words (e.g. bookshop) can also be presented in this way.

A word’s frequency

Knowing how frequently words occur in speech or text can help decide how often to use a word and the appropriate register level. Applying this knowledge accurately may help a learner sound more native-like, and conversely using it inappropriately may have an uninspiring affect. With the help of corpora data it is possible to get detailed information on how frequently words occur, and using this as a benchmark, the quality of a learner’s frequency intuition can be assessed.

Research in this area has been limited and the results inconsistent. Ringeling (1984), for example, found that it is possible for advance L2 learners to reach native-like frequency intuition levels. Schmitt and Dunham (1999) concluded that education influences the extent of a person’s frequency knowledge. Aizawa, Mochazuki and Meara (2001) demonstrated that there is little difference between intuition levels of native and non-native speakers, and that a native speaker’s intuition often disagrees with corpus data. The implication of this last point is that teachers need to rely more on corpus data in the classroom. In the most recent study, McCrostie (2005) presented 20 native and 105 non-native subjects with 3 tests, asking them to rank sets of words according to frequency. The results echoed the previous findings with the exception that education was found to have little effect on frequency judgments.

Factors such as social background, education, age, location and context may have a considerable affect on a person’s estimate of a word’s frequency. These factors plus the need to use context relevant corpus data should be an important consideration for future research.

A word’s register

The affective or connotational meaning of a word adds to or changes meaning depending on the function or situation (Richards 1976). Two words may refer to the same object or action but the overtones attached to its use may be different (Palmer 1976). People within a society or among different cultures often have different associations with certain words, which can also result in different interpretations. The way we use this hidden meaning of a word, referred to as register, can have several variations. One attempt to categorise these variations was by Chui (1972), see Table 5.

Table 5: Register variations (Chui, 1972)

Temporal Variation

words can be in or out of fashion.

Geographical variation

different words may be used for the same item depending on geographical location.

Social variation

social class can influence word selection.

Social role variation

level of formality depends on social relationship with the addressee.

Field of discourse variation

the topic of discussion influences language choice.

Mode of discourse variation

whether discourse is written or spoken affects word choice.

Teaching word meaning in the complete sense is obviously a challenging task for any teacher, and careful planning is needed to include register in a lesson plan. Likewise, it is also difficult for a learner to absorb all aspects of word meaning in one encounter. Starting with the most common conceptual meaning of a word, other aspects of meaning should be introduced gradually with subsequent encounters (Schmitt 2000). Learners need to be exposed to register use in real contexts, relevant reading and listening activities can be a useful way of doing this (Gairns and Redman 1986).

A word’s grammar

Grammatical knowledge of a word is key to the accurate use of a word in communication. The main aspects of this knowledge that I will highlight are word class, morphology and the more recent, grammatical patterns.

The importance of word class knowledge is supported by the idea that with improved learner proficiency there is a shift from syntagmatic to paradigmatic associations, something we looked at earlier. It could well be that this transition is aided by a developing word grammar knowledge. However, this knowledge seems less important when considering that a majority of native speakers do not know the class of a word. In a study involving French and Linguistics students at a British university, Alderson et al (1997) found that most native speakers were unable to match a word (with the exception of nouns and verbs) to its correct class, even though they were aware of the meaning. One reason of course, could be that that they lacked knowledge of the relevant terminology. In a similar study by Odlin and Natalicio (1982) with native speakers (NS) of English and Brazilian ESL students (intermediate to advance levels), the NS scored just over 80% and the highest ESL student score was 75% in a word class identification test, showing that it is possible for learners to gain native-like proficiency in this area.

Another aspect of word grammar knowledge, morphology, involves using the correct affixed form of a word. For example, when a context requires a verb, help is needed and when an adjective, it is helpful. An awareness of the connection between the parts would most likely facilitate the acquisition of the different forms (Schmitt, 2000). In fact, the rapid lexical acquisition of L1 speakers of English in childhood most probably relates to a developing ability to recognise the morphological parts of a word (White et al, 1989). So, for example, a student who has learnt the word adore might be able to comprehend adorable, adorer and adoration through an understanding of the constituent morphemes.

With regards to how the different morphological forms are stored in the mental lexicon, Nagy et al (1989) looked at how the frequency of individual word forms and the combined frequencies of word family members influence word recognition speed. They asked 95 US college students to separate stem from non-stem words in a lexicon decision task, and found that the frequency of inflectionally and derivationally related words had a considerable affect on the speed and accuracy of recognition of stems. Their results support the idea that words are connected in the mental lexicon by both derivational and inflectional relations.

In order to produce a word accurately and fluently, knowledge of the grammatical constraints surrounding the word is essential. The word has to occur in a finite number of possible arrangements. After examining the 250 million word Bank of English corpus at Cobuild, Hunston et al (1997) found that every word can be defined in terms of the patterns that it typically occurs in. The verb eat, for example, occurs in two patterns: on its own ‘He ate’, or it is followed by a noun group ‘He ate a banana’. Similarly, they argue each word has an associated pattern and it is these patterns which combine both vocabulary and grammar to form the building blocks of language. Analysing concordance data from relevant corpora would be an excellent way of highlighting these patterns in the classroom.

Learning the morphological connections between words requires, as a prerequisite, knowledge of word classes. So from a pedagogical perspective it is important to present learners with word class information when teaching a word. Although inflectional forms of a word can be acquired implicitly by analyzing its parts, derivational forms are not necessarily as transparent. Results from a study by Schmitt and Zimmerman (2002) found that having knowledge of one word within a word family does not imply the same with other members of that family. Hence, even with derivational forms, explicit attention is probably necessary.

A words collocations

Knowledge of a word’s collocations, the way it combines with other words in natural communication, is another step towards fully knowing a word (Nation, 2001: 317). A collocation can occur in a variety of forms but broadly speaking, the term refers to a common grouping of words to form phrases or clauses which may not be fully semantically transparent. The link between the mental lexicon and syntagmatic associations, which we looked at earlier, justifies the relevance of collocations to the language acquisition process.

The process of producing collocations can be explained by looking at two types of language production suggested by Sinclair (1991). The first is the idiom principle, where language is seen as consisting of pre-fabricated chunks, and the second is the open-choice principle which sees language creatively producing new combinations of words. In his research DeCock et al (1998) observed that there was only a minor difference in use, of the idiom principle, between native and non-native subjects. However the chunks used by the learners seemed to be less varied. In a study of written texts by native and advanced non-native students Howarth (1996) found that collocation density (restricted collocations and idioms) for natives was 13% higher, which suggests that development stages may exist for collocation use.

To gain proficiency in collocation use requires a lot of learning, too much for it to be covered effectively in most classrooms. Hence, it is important to direct learners gradually towards more autonomy. At the beginning stages, frequent or useful collocations can be memorized as unanalyzed chunks (Nation, 2001). At more advanced stages students could be taught to work with concordances and analysing contextually relevant corpus data.

Conclusion

In this essay I outlined some of the knowledge features required to comprehensively acquire a word. The ideas were broadly structured around the framework presented by Nation (1990). Some of the ways teachers can assist in the learning process were also discussed.

I would like to conclude this discussion by reflecting on the immense burden that word knowledge, in its entirety, places on learners. It can take up to 20 encounters before all features of a word can be considered known (McCarten 2007). Consequently vocabulary should be taught gradually and an effort made to make it personal and motivating in the classroom. With an even greater priority, learners should be encouraged to seek more exposure through autonomous learning opportunities.