Skip to navigation – Site map

What's in a Word? Conversion vs Construction

Walter Hirtle
p. 7-20


Dans la langue anglaise, le problème posé par des emplois tels que Will he medal tonight? (commentaire entendu pendant les Jeux Olympiques) ou Is better always good? est souvent traité comme le résultat de la « conversion » d’un nom en verbe, d’un adjectif en nom, etc., un procès qui présuppose que les mots, prêts à employer dans une phrase, sont stockés dans une sorte de lexique mental. On peut mieux expliquer de tels emplois si on adopte, comme point de départ, la perspective traditionnelle selon laquelle le signifié d’un mot comporte un élément lexical et un élément grammatical, et que ces deux constituants sont assemblés chaque fois qu’on a besoin d’un mot pour le discours. Afin de découvrir ce qui permet la création à volonté de mots nouveaux, comme dans Petruchio is Kated de Shakespeare, on partira du postulat que les ressources d’une langue fournissent au sujet parlant non des mots toutfaits, mais plutôt les éléments formateurs des mots.

Top of page

Full text

1. Introduction

1A few years ago during the summer Olympics, I heard a commentator on TV ask a question concerning an evening event:

Will he medal tonight?

2Here medal, common as a substantive, functions as a verb, an infinitive. Is it a different word? More intriguing is the following citation picked up by an observant student from a university publication:

I see that kids are focused on science. They’re asking science questions. They’re ‘sciencing’ as (CETUS researcher) David Blades says. (UVIC Torch, autumn 2006, p. 24)

3Here we find the same word used in three different ways—or are there three different words? Another perceptive student picked up the following sentence spoken by a teenager on the radio:

They recently unemployed 500 people.

4The example is intriguing because it raises the question of how unemployed, usually used as an adjective (or a substantive), came to be used as a verb (though it might be argued it was derived by means of a prefix from the verb employ).

5Such examples are not limited to occasional uses of verbs which strike us as original. Probably She’s so Hollywood would not sound unusual to most people and everyone remembers the book title Small is Beautiful. As the title of an article Is better always good? certainly catches the attention. Many teachers would likely say that the adjectives small and better, being the subject of the sentence, are used as nouns here but might hesitate to say that the proper noun Hollywood is used as an adjective. In any case, the same question arises: whether an adjective can be “used as” a noun and a proper noun as an adjective, as some grammarians put it (cf. Curme, p. 534-8; Schibsbye, p. 123-8; Christophersen and Sandved, p. 115-7), or whether these are “totally or partially converted” into another part of speech (Poutsma, p. 792; cf. also Zandvoort, p. 265-77), i.e. different words expressing much the same idea.

6The novelty of some of the above uses should not however lead us to think that the process permitting them is of recent invention. When we recall lines like

I warrant him, Petruchio is Kated.


Shall sweet Bianca practice how to bride it? (Taming of the Shrew (III, ii, 245 and 251))


A mile before his tent fall down and knee
The way into his mercy.


Nay, [he] godded me indeed.

from Coriolanus (V, i, 5-6 and V, iii, 11) we realize that this capacity to come up with a new use, or new word, even for nonce uses has long been available to speakers of English and enables the imaginative speaker/writer to achieve remarkable effects.

7From a linguistic point of view, the interesting thing about such examples is not so much the effect they produce but the processes involved in producing them. How is it that speakers can invent a new use? word? on the spur of the moment in this way, inventions which, if generally adopted (cf. a go and a think), become so commonplace that they are no longer noteworthy? Words like use, process, produce, effect, to take examples from the last two sentences, illustrate what grammarians call “conversion” in this or other uses, though use, process and effect as verbs, and produce as a noun would probably not be felt as “converted” today. Even for such words, however, at some point in the history of English a speaker had to innovate, and it is this process which deserves attention. The linguist should, like Quirk et al. (1558), “treat conversion not as a historical process, but rather as a process now available for extending the lexical resources of the language”.

2. Conversion

8The first step in exploring this process is to examine the term “conversion” itself to see how it is understood. For Quirk et al. (1558) “Conversion is the derivational process whereby an item is adapted or converted to a new word class without the addition of an affix”. As such, it is “unusually prominent as a word-formation process”. For Huddleston and Pullum (1640) “Conversion normally involves changing a word’s syntactic category without any concomitant change of form… We include conversion within the set of lexical word-formation processes because we see it as creating new words”. These grammarians thus agree that this process produces a new word, not an old word in a new use, and the latter grammar (1640) gives the reason for this: “we regard any difference in primary category as sufficient to establish a difference between one word and another”. Neither the expression “primary category” nor the above “word class” will be adopted here for “part of speech” since the traditional expression refers us to the reality of what a speaker does rather than what grammarians establish: speakers produce words that are distinct functional parts of the sentences making up speech or text. Granted this difference in terminology, however, the important point is that both of these major grammars distinguish words by their part of speech: medal used as a noun and medal used as a verb are different words and not merely different uses of the same word. This is a crucial starting point because it presupposes that the part of speech is an essential component of a word’s formation in English and not merely a set of observable functions. Moreover, this distinction of grammatical form within the word is not signified morphologically by any affix or change of form but only syntactically, usually by positioning in the sentence.

9The next step in this attempt to discern how the two words differ in order to work back to the process that produced the new one brings us to a statement by Huddleston and Pullum (1641): “It is a notable property of English that it has a great deal of homonymy between nouns and verbs”. To consider medal noun and medal verb homonyms is, to say the least, surprising since we customarily consider homonyms to have like signs but different meanings. Granted that between noun and verb there is a difference of meaning, this is not what is felt in a typical case of homonymy like pool (of water vs. the game). Pool confronts us with quite unrelated ideas, different lexical meanings, whereas medal confronts us with distinct parts of speech, different grammatical meanings. That is to say, homonymy as usually understood involves distinct lexical meanings whereas the problem we are discussing here involves distinct grammatical meanings. Thus while one can hardly consider these words to be homonyms, the above statement does help to bring out a very important fact: the meaning of a word is binary, consisting of two types of meaning-import, lexical and grammatical, and according to the grammarians what is involved in the word-forming process discussed here is a change in the latter, the part of speech.

10A final comment on the grammarians’ view of the process involved concerns the term “conversion” itself. This term presupposes that one word is changed into another, for example “the creation of the verb humble from the adjective humble. That is, “a word is formed from a pre-existing morphological unit by simply giving it new grammatical properties”. (Huddleston and Pullum 1641) In the light of what we have just seen, this would appear to involve stripping the lexeme ‘humble’ of its grammatical meaning, its part of speech adjective, and forming it with a new one, the part of speech verb, a process which is anything but obvious. “Simply giving” a word a new part of speech would involve de-forming the word and re-forming another word, a process which remains hypothetical, not to say mysterious, until the parts of speech involved are described in terms of their different grammatical meanings and how the word’s lexeme is endowed with these grammatical properties. An idea of the complexity implied by the notion of “converting” one word into another can be gained by the listing in Quirk et al. (1560-3) of deverbal nouns, de-adjectival nouns, denominal verbs, de-adjectival verbs, denominal adjectives and other “minor categories”.

11These considerations point to the underlying difficulty here. The idea that one word is converted into another is based on the presupposition that a word is stocked in our memory like a sandwich in a dispensing machine: the speaker simply has to punch the right button and out it pops ready-made. As one study (Kosslyn and Koenig, 211) puts it: “A normal speaker produces about three words a second. These words are extracted from a stored mental dictionary (a lexicon) of somewhere between 20,000 and 50,000 words”. This conception of words as preconstructed entities is that of the ordinary speaker since words actually do emerge into consciousness fully formed and ready for use in the sentence, and so one can understand why grammarians take it for granted. Even linguists however appear to accept the same presupposition when they include words in the lexicon, the storehouse of “lexical entries”, along with prefixes, suffixes and other elements they have not been able to analyze. Since our common experience of language, or anything else, does not always provide a faithful reflection of the whole of reality, one would expect a linguist accepting this assumption at least to question it, and yet, to my knowledge, none have. After all it is the scientist’s task to examine any such presupposition when it fails to explain certain observations, like the various “conversions” listed in Quirk et al. To gloss over them as “simply giving” a word another part of speech cannot be considered an explanation. This is clearly illustrated by an example like round, often cited for its versatility since it is readily formed by five parts of speech. Should round be considered a single lexical entry as, say, an adjective in the subconscious lexicon with four possible conversion processes—one to form it as a noun, another to form it as a verb, another to form it as an adverb, and another to form it as a preposition—each of which remains to be described? Or should it be considered as five homonyms, five separate lexical entries? Or is there another way of understanding what is going on here?

3. What’s in a word?

12In order to propose another way of conceiving the word-forming process we will consider things from a more general point of view by asking: What’s in a word? This will lead us first of all to bring in certain historical facts and to envisage words not just in English but in the Indo-European languages as a whole. Our discussion will not take into account the sign, the physical component of a word, but only the meaning, the mental component of a word.

13The historical considerations are presented by Michael (44-7) in his study of English grammars up to 1800. He points out that the word was defined in antiquity as “the smallest unit of discourse”. Later, in the middle ages, the speculative grammarians tried to probe the makeup of the word. They “base their analyses on a threefold distinction between vox, the mere speech-sound; dictio, the word regarded as a meaningful speech-sound; pars, the word regarded as a syntactical unit”. This led them to propose that a word signifies directly its notional import or lexeme and signifies indirectly, “consignifies”, its grammatical import, its part of speech. Since it is the part of speech which determines a word’s function in the sentence, “the syntactical function of a word is part of its meaning”. The interesting thing here is that the distinction between lexical and grammatical meaning, between signifying and consignifying, implied by the contemporary grammarians cited above had already been discerned by grammarians in the middle ages. On the other hand, Michael tells us, “The renaissance grammarians made no use of the two most important ideas about the word which were available to them: Dionysius Thrax’s description of it as a minimum unit of discourse and the speculative grammarians’ distinction between semantic and syntactic units”. Even today these two ideas tend to be neglected. As a consequence there is little discussion of words as the fundamental linguistic form, the necessary constituents of every act of language, and, in fact, some linguists even deny the word’s existence (cf. Mounin, 222). Furthermore, many contemporary linguists consider the mental makeup of words, like that of affixes, to be unanalyzable, and so for the most part there are only haphazard attempts to analyze the relation between a word’s grammatical and lexical meaning imports.

  • 1 It would be necessary to search for the reason for dividing language into words—for in spite of th (...)

14This may seem surprising for anyone familiar with languages where the declension or the conjugation are more extensive than in English because the grammatical inflexions make it obvious that a given lexeme can be grammatically formed in various ways. Moreover in the 19th century development of historical grammar based on the comparative method, this distinction was observed on the level of the sign and could have provided a basis for analyzing the mental makeup of words. Saussure (154) recognized that « Il faudrait chercher sur quoi se fonde la division en mots – car le mot, malgré la difficulté qu’on a à le définir, est une unité qui s’impose à l’esprit, quelque chose de central dans le mécanisme de la langue; –mais c’est là un sujet qui remplirait à lui seul un volume».1 To my knowledge, Gustave Guillaume, perhaps sparked by this very passage from the Cours, was the only linguist who took up the challenge of analyzing the word as the central problem in linguistics. In 1957, he explained “[h]ow, almost half a century ago, I saw the link between semiological observation (as ordinarily found in comparative linguistics) and psycho-systematic observation, which I was soon to consider devoting a lifetime to” (1984, 39), a career which has resulted in the posthumous publication of over 20 volumes so far.

  • 2 Except for the 1984 volume, all citations from Guillaume have been translated by myself.
  • 3 Tongue” is used here in the sense of “the power of communication or expression through speech” (We (...)
  • 4 To contrast with the potential of tongue, Guillaume replaced Saussure’s parole, speech, by discours (...)

15Guillaume’s first volume (1919), on the article in French, begins as follows: “The present work is an essay applying the comparative method to the formal [grammatical] part of languages”.2 (11) To adopt a method that had proven its worth in historical studies of the sign to a synchronic study of grammatical meaning involved a challenge which his volume did not meet, although it did pose the problem squarely. It was only some twenty years later, in the early forties, that Guillaume finally discerned the mental system (or as he called it, the psychosystem) of the article. This breakthrough arose within the framework of a far more general challenge which he took up in his 1941-42 lectures (B series): “The subject I am treating this year, which is the limitation and the construction of the word, has never to my knowledge been examined by anybody from the point of view I am adopting”. (2005, 53) His starting point was the observation that speakers do not improvise the means of expression when they speak. This led him to conclude that every language provides the means required to construct the words needed for producing the sentence(s) expressing whatever a speaker has in mind at that moment. Moreover these means are a permanent, systemically organized resource in the mind permitting a speaker to express an endless variety of messages. As such, one’s language exists as a potential—what he calls la langue, tongue3—for forming words ready to play their part in a sentence, which is the unit of actualized language—what he calls discours,4 discourse. Thus Guillaume focuses on the word, which is both the output produced by the preconscious mental processes of tongue and the input permitting the syntactic processes producing discourse.

16What he seeks then is the psycho-system of the word in any language, a variable from one language type to another. He proposed as the basic schema of word-formation for the Indo-European languages a binary process whose first phase involves representing whatever entity, process, etc. one has in mind to talk about by means of a lexeme, a particular notion belonging to one’s acquired vocabulary. He does not speak of all the notions in tongue as entries in a lexicon since this would imply a stock of already formed words, “pre-existing morphological units”, in the memory. Rather he speaks of them as forming an “idea-universe” to suggest that all the ideas or notions acquired are ready to categorize and represent whatever may arise in a speaker’s momentary experience of the universe. This first phase produces what a word signifies, a specific idea differentiating a given word from all others. This is what, in spite of a common sign, makes pool, the game, a different word from pool, of water. The first phase, which Guillaume calls ideogenesis, thus consists of a particularizing operation since it results in discerning a particular lexeme.

17The second phase of the word-producing process consists of forming this lexeme grammatically, categorizing it by means of appropriate morphemes to produce the part of speech required for the word to fulfill the syntactic role foreseen for it in the intended sentence. This second phase consists of a generalizing operation since it results in what a word consignifies—a part of speech common to many other words. It is not possible here to give a description of what is involved in the morphogenesis of a verb, forming the lexeme by means of aspect, mood, tense, etc., or for the noun, whose lexeme is categorized for gender, number and case (cf. my 2007 studies, a and b respectively, for the detail). This mental process of giving a general form to the lexical matter discerned in ideogenesis Guillaume called morphogenesis. He often represented this binary process for forming words in the Indo-European languages as in the following diagram:

4. Constructing vs. converting

18Guillaume’s general theory enables us to postulate that this is how speakers of English form a word, every word, while speaking or writing. The main point to be made in presenting this theory of word formation is that the processes for categorizing the lexical matter are different for each part of speech, and this brings us back to the use of medal as a verb in the example given above. Each part of speech has its own formative subsystems without which it cannot provide the general form required for a word to take its place as a syntactic unit in the sentence being constructed. Thus to be able to situate an event in time and attribute it to a support, the time involved and the relation to the subject must be represented by aspect, mood, tense, voice and person. Each of these subsystems provides a process of representing one of the grammatical components of a verb. On the other hand, to situate an entity in space as a support, the space involved must be represented by gender, number, case and person. Each of these subsystems provides a process of representing one of the grammatical components of a substantive. That is, although the operation of morphogenesis is common to the forming of every type of word, the processes involved in it vary from one part of speech to another since they arise from the subsystems specific to each part of speech. So the lexeme ‘medal’ when called on to be used as a verb is not categorized, for example, by the system of gender, just as the same lexeme, when called on to serve as a substantive, is not categorized by the system of tense.

19From this it follows that each part of speech provides a word with a different syntactic capacity. A verb can function as a predicate because it has been formed grammatically as a time word, whereas a substantive can function as a subject or direct object, etc. because it has been formed as a space word. That is to say, the syntactic relations a word can establish with other words in a sentence are determined by its grammatical makeup, its part of speech—what it consignifies, in the terminology of the medieval grammarians. Humboldt (128) expressed the same idea differently: “Languages like Sanskrit… already weave into the unity of the word its relations to the sentence”. According to Valin (viva voce), Guillaume often pointed out that “a language has, in the final analysis, the syntax of its morphology”. (2007, 10) The implications of this for studies of syntax are important but cannot be explored here.

20Granted this view of a word, it makes little sense to speak of “simply giving it new grammatical properties”, as though a word’s grammar were something like a coat one puts on over the lexeme. Rather, a word’s grammatical makeup configures its lexical matter. Without this configuration, a lexeme cannot enter into a sentence, be a part of the speech, of the discourse, a speaker is uttering. It cannot even be called to mind. A word’s grammar is not simply added on to its lexical import but rather informs it, determining the word’s nature so that it can function in the way foreseen for it. If then it is granted that the relation between a word’s two mental components is established while the act of language is going on—during the split second the word is being constructed—the idea of “converting” one word into another appears to be a stopgap based on the way grammarians compare two different uses and not on what a speaker does while speaking. Hence it contributes little to our understanding of language as a phenomenon.

21It might be thought that to view a word in this way is useful for explaining uses in innovative examples like the above, but not for ordinary uses. This would ignore the fact mentioned above (and illustrated again by thought, view, word, above, ordinary in the preceding sentence) that a great many lexemes are configured by different parts of speech without our even being aware of it and so the problem is far more general than one might think. This would also ignore that Guillaume’s conception of word construction is general enough to include all words that can establish syntactic relations in a sentence, and that is, to my mind, an important achievement.

5. How ‘medal’ “got verbed”

22It remains to suggest the way a lexeme like ‘medal’ is treated so that it emerges into discourse as a verb. This involves outlining the way any word is constructed during the present of speech, with the particularity that here it involves an innovation. Within the context of the Olympics, medal usually designates the gold, silver or bronze object won by the best athletes. Like any other notion in tongue, ‘medal’ is part of that inner universe consisting of what Guillaume calls “viewing ideas”, ideas constantly on standby to categorize whatever arises in our stream of consciousness. Having in mind a particular swimmer’s performance as compared with that of the other competitors in the coming event, the speaker represented it by means of the notion ‘medal’, which, as a potential lexeme with no grammatical strings attached, must be actualized lexically and formed grammatically. Because in that experiential situation the metal object constitutes the purpose of the swimmer’s performance, it is necessarily linked to the swimming and so, in order to express the activity, the speaker can grammaticize ‘medal’ as a verb with the assurance of being understood. Confronted with an infinitive, listeners are constrained to look for a process presupposed by the lexeme and so, even though it may be a new use for them, following this link from the object to a process seen as a pre-condition leads them to what the speaker had in mind, to what is targeted in the intended message by the verb medal, i.e. the subject’s performance in the race.

23There are of course various accidental circumstances which could be taken into account here, whatever momentary impressions arising in the experiential situation that contribute to the speaker’s intended message. This applies to listeners as well since some might well consider that there is another process necessarily linked with the metal object, namely the awarding ceremony, and so interpret the sentence as meaning ‘Will he receive a medal this evening?’ This possible ambiguity between the performing and the awarding in no way obstructs the commentator’s communication but it does help confirm the analysis just presented since both interpretations are based on what is seen as a necessary link with the commonplace understanding of ‘medal’.

24Worth noting also is the fact that this use of medal appears to be limited to sporting situations. It would hardly be found in, for example, the situation of a soldier going into battle since winning a medal is not the purpose of such an activity. The link between the object and the activity is not the same as in sporting events.

25These accidental circumstances help to bring out the potentiality of the lexeme as a viewing idea in tongue ready to represent whatever it focuses on in the speaker’s experience. When called on during the act of languaging a momentary experience, a lexeme can provide only a highly abstract representation of an object or activity or quality but all the accidental circumstances involved, both linguistic (the grammatical form it is given in the word, the impact of other words in the sentence, etc.) and extra-linguistic (the message arising from previous discourse, knowledge of the particular circumstances and general situation shared with the addressee, etc.), permit the speaker to leave implicit elements required to get to the specifics of the intended message without impeding the intended interpretation of the sentence by a listener aware of these circumstances. The important point for our discussion is that a lexeme in tongue lends itself to different interpretations in discourse. The fact that, before being used, it is free of any grammatical attachments entails that the particular form given a lexeme by the speaker through morphogenesis helps determine how it will be interpreted in the sentence. Making the part of speech an inherent part of the word contributing to what it expresses is a far cry from simply “converting” a substantive into a verb.

26Does this understanding of how a speaker produces a new word in the case of medal apply to other cases? For the second example above:

I see that kids are focused on science. They’re asking science questions. They’re ‘sciencing’ as (CETUS researcher) David Blades says,

27assuming that ‘science’, like ‘medal’, is a viewing idea in tongue would suggest that here too some activity is presupposed, and indeed one interprets the sentence to mean ‘they are applying the scientific method in some way’. That is, even a nonce use like this one is readily understood because by configuring the unchanging potential lexeme as a verb, the speaker induces the listener to find the process it implies. This of course contrasts with its use in the first sentence, where the lexeme is configured as a substantive. The same applies to the second sentence, where it is configured as an adjective, a possibility presupposing that the general notion ‘science’ implies certain characteristics or properties of questions asked. The fact that one would expect scientific for the adjective and doing science for the verb helps bring out the point here: the lexeme in tongue is malleable, ready to be formed through morphogenesis.

28Another aspect of the question is raised by the third example:

They recently unemployed 500 people.

29Assuming that ‘unemployed’ is a notion in tongue, to configure it as a verb presupposes that the quality or state it usually represents is linked to some process, that leading to the state, i.e. being dismissed from an employment. On the other hand, ‘employ’ might be considered the notion in tongue from which ‘unemployed’ in the example was produced by derivation through the lexical augment signified by the prefix, its configuration as a verb being consignified by the suffix. Whether or not ‘unemployed’ is considered a lexeme instituted in tongue, or one which is still derived by the processes of word formation during the act of speaking probably depends on the individual speaker.

30The next three examples:

She’s so Hollywood.
Small is Beautiful.
Is better always good?

31involve the relation between substantives and adjectives. Again it would be hard to imagine how one can be “converted” into the other, but envisaging these examples from the point of view of how the lexeme is categorized grammatically provides a readily understandable explanation of the resulting use. Thus, forming ‘Hollywood’ not as a proper noun representing a particular place or industry but as an adjective has the effect of reducing its notional import to the characteristics or qualities one associates with that place or industry (and this import is further reduced in the sentence to characteristics which can be attributed to the subject, a human person). ‘Small’ on the other hand is configured as a substantive, which has the effect of the lexeme expressing not only its usual characteristics pertaining to size but also that to which these characteristics are applied so that we understand something like ‘whatever is small…’ ‘Better’ is also grammaticized as a substantive but the result in discourse is not quite the same as in the preceding sentence since we understand not just the qualities applied to something but rather ‘trying to improve something’. The suggestion of an activity arising here may well be the result of the comparative treatment of the qualities evoked, but this can only be confirmed once the role of the -er suffix with adjectives has been analyzed.

32The examples from Shakespeare help bring out the role of non-linguistic factors in prompting inventions of new words. Expressions like to bride it, knee the way into his mercy and [he] godded me might well be understood out of context since there is an essential relationship linking ‘marry’ to ‘bride’ and ‘crawl’ to ‘knee’ and ‘deify’ to ‘god’. However without some idea of the characters named Petruchio is Kated would hardly be comprehensible since the relation between ‘marry’ and ‘kate’ is limited to the persons involved in that particular situation. That is to say, innovations like these, which all arose some 400 years ago but may be new for many readers today, are the result of calling on an already instituted lexeme to evoke a relation newly perceived in the particular situation the speaker/writer has in mind and so to express the other term of that relation—here an activity or its outcome. In this respect, the capacity of lexemes to lend themselves to this “process now available for extending the lexical resources of the language” gives rise both to striking nonce uses and, over time, to more permanent common uses of a lexeme, as illustrated by round.

6. Conclusion

33The assumption that the words in our examples are items “stored in the speaker’s mental lexicon”, as Taylor (2002, 74) puts it, leads either to considering that they constitute a set of homonyms—hardly consistent with our concept of homonym—or to the idea of conversion. The argument developed above that medal verb is not derived from medal noun runs counter to the attitude of the ordinary speaker (as shown in the last example below) and so may well be “difficult to envisage”, as one reader has put it, but unless the processes of de-categorizing and re-categorizing are described within the framework of how words in the lexicon got grammatically categorized in the first place, the usual manner of envisaging the phenomenon is, to say the least, not convincing.

34On the other hand, to confirm the approach proposed here would require a confrontation with more detailed studies of this innovative potential of speakers. Because examples like those discussed here are innovative, the general pragmatic or other linguistic conditions which might incite a speaker to exploit this potential of a particular lexeme for the first time are not apparent. The fact that this innovative possibility is exploited initially by an individual bears witness to the speaker’s sentence intent (visée phrastique, in Guillaume’s terms) to form a particular notion as a specific part of speech, as in Shakespeare’s But me no buts. Furthermore, someone has proposed that this view of word formation might be accepted for “stylistically marked” cases but not for other “lexical units”, but this would ignore the far more extensive data base consisting of innovations that are now part of history: is a case like the verb base which, having lost its innovative stylistic mark, we take for granted today, to be considered a homonym of the noun or a recategorization of it? It is more satisfactory to consider that those cases of innovation we are aware of evidence the need to categorize any lexeme for it to be used in a sentence and so they support the view that all words are constructed each time we use them.

35The assumption that words are formed in the speaker’s mind during the moment of speaking or writing leads to distinguishing between successive phases in their construction, ideogenesis and morphogenesis. This approach allows for formatting a lexeme the same way repeatedly or in different ways, or even in a new way, according to the requirements of representing and expressing the individual experiential content constituting each intended message. It is based on the postulate that we have something far more useful than a mental lexicon as a permanent resource in our preconscious minds, namely the potential for producing words. That is, we have the lexemes and the grammatical systems with their mental programs or mechanisms for constructing the words required in the moment of need. This fluid manner of imagining words during the act of speaking or writing implies that every word we construct lasts no longer than the sentence it becomes part of, but that the means of forming it, acquired with our mother tongue, are an indelible resource of the mind.

36The problem of formatting a lexeme as different parts of speech can thus be explained within the framework of a general theory which distinguishes between the operations providing the two mental components of any word. As is often the case when focusing on a particular question within the framework of a general theory, explaining one problem throws light on another one. One such problem was raised by Patrick Duffley (private communication): how to analyze an example involving a noun phrase like She’s so last century, where the incidence of ‘last’ to ‘century’ has already been accomplished. This pertinent question might be explored in the light of what Valin (1981) calls a “proto-phrase” (proto-syntagme), a phrase still in the making since the substantive century has not yet been made incident to its own extensity. Unfortunately this lead cannot be developed within the bounds of the present article.

37Here, we have touched on the functioning of lexemes as potentials in tongue, suggesting that in an innovative use the speaker establishes a new relationship thanks to the lexeme and, through the word’s grammatical relationships in the sentence, induces the listener to do likewise. This capacity of the lexeme to represent something through a different relation can perhaps be understood better in the light of Guillaume’s proposal that lexemes are “viewing ideas”, notions instituted in tongue unconsciously categorizing whatever arises in our stream of consciousness whether or not we decide to talk about it (a role recognized in human perception by certain psychologists). Thus envisaging how a swimmer will perform in the final was automatically categorized in terms of ‘medal’, the goal, but this had to be grammaticized as a verb to bring out the process of achieving it (or the resulting ceremony), and not the goal itself.

38The point here is that Guillaume proposes that lexemes play a role in perception, an active role that may well help to explain historical development of meanings. Such developments have to start somewhere, initiated by individual speakers discerning a new link between the lexeme as instituted in tongue and something in their momentary experience. If, as in the recent case of ‘access’ being formed as a verb, the innovation is picked up by other speakers it may lead to a generalization of the lexeme. Although these considerations concern the field of lexical semantics, a field which in English has received little attention within the theoretical framework adopted here, the type of example discussed above does provide an opportunity to observe what happens when speakers, and as a consequence listeners, innovate. The following exchange between two comic strip characters commenting on language will conclude our discussion and give readers the opportunity to construct new words:

Access got verbed”.
“I like to verb words”.
“Verbing wierds language”.

Top of page


Christophersen, Paul and Arthur O. Sandved. 1969. An Advanced English Grammar. London: Macmillan.

Curme, George. 1931. Syntax. Boston: Heath and Company.

Guillaume, Gustave. 1919/1975. Le problème de l’article et sa solution dans la langue française. Paris: Nizet; Québec: Presses de l’Université Laval.

Guillaume, Gustave. 1984. Foundations for a Science of Language. Amsterdam and Philadelphia: John Benjamins.

Guillaume, Gustave. 2005. Leçons de linguistique de-, vol. 17: 1941-1942 B, Théorie du mot et typologie linguistique: limitation et construction du mot à travers les langues. Québec: Presses de l’Université Laval.

Guillaume, Gustave. 2007. Essai de mécanique intuitionnelle I: Espace et temps en pensée commune et dans les structures de langue. Québec: Presses de l’Université Laval.

Hirtle, Walter. 2007a. Lessons on the English Verb: No Expression without Representation. Montreal: McGill-Queen’s University Press.

Hirtle, Walter. 2007b. Language in the Mind: An Introduction to Guillaume’s Theory. Montreal: McGill-Queen’s University Press.

Huddleston, Rodney and Geoffrey Κ. Pullum. 2002. The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press.

Humboldt, Wilhelm von. 1836/1988. On Language. Translated by Peter Heath.

Cambridge: Cambridge University Press.

Kosslyn, Stephen M. and Olivier Koenig. 1992. Wet Mind: The New Cognitive Neuroscience. New York: The Free Press.

Michael, Ian. 1970. English Grammatical Categories and the Tradition to 1800.

Cambridge: Cambridge University Press.

Mounin, Georges. 2004. Dictionnaire de la linguistique, 4e edition. Paris: Presses Universitaires de France.

Poutsma, H. 1926. A Grammar of Late Modern English, Part II, Section II Groningen: P. Noordhoff.

Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. 1985. A Comprehensive Grammar of the English Language. London and New York: Longman.

Saussure, Ferdinand. 1916/1955. Cours de linguistique générale. Paris: Payot.

Schibsbye, Knud. 1970. A Modern English Grammar, 2nd ed. London: Oxford University Press.

Taylor, John R. 2002. Cognitive Grammar. Oxford: Oxford University Press.

Valin, Roch. 1981. Perspectives psychomécaniques sur la syntaxe. Québec: Presses de l’Université Laval.

Webster’s Third New International Dictionary. 1969. Springfield, Mass.: Merriam.

Zandvoort, R. W. 1957. A Handbook of English Grammar. London: Longmans Green.

Top of page


1 It would be necessary to search for the reason for dividing language into words—for in spite of the difficulty of defining it, the word is a unit that strikes the mind, something central in the mechanism of language—but that is a subject which by itself would fill a volume”. (Translated by Wade Baskin, 1959, New York: Philosophical Library) For «une unité qui s’impose à l’esprit» I would prefer “a unit which imposes itself on the mind”, and for «le mécanisme de la langue», “the mechanism of tongue” (see below, note 3).

2 Except for the 1984 volume, all citations from Guillaume have been translated by myself.

3 Tongue” is used here in the sense of “the power of communication or expression through speech” (Webster’s s.v.) as in “the mother tongue” (we do not say *the mother language). Those not used to this sense are referred to the Introduction (p. xx) in Guillaume 1984 for a discussion of the problem of translation involved here. The point is that language cannot bring out the contrast of langue + parole. Moreover, the French term langue when used in English has a purely static sense because it is associated with Saussure’s «linguistique statique» (p. 154) of synchrony, whereas this uncommon use of tongue can suggest what characterized Guillaume’s view of language-as-a-potential, namely the dynamism inherent in any potential.

4 To contrast with the potential of tongue, Guillaume replaced Saussure’s parole, speech, by discours, since the former has both a potential existence (cf. phonology) and an actual existence (what we hear), whereas discourse exists only as actual sentences.

Top of page

List of illustrations

File image/jpeg, 27k
Top of page


Bibliographical reference

Walter Hirtle, « What's in a Word? Conversion vs Construction », Anglophonia/Sigma, 11 (22) | 2007, 7-20.

Electronic reference

Walter Hirtle, « What's in a Word? Conversion vs Construction », Anglophonia/Sigma [Online], 11 (22) | 2007, Online since 13 December 2016, connection on 27 June 2017. URL : ; DOI : 10.4000/anglophonia.1083

Top of page

About the author

Walter Hirtle

Fonds Gustave Guillaume – Université Laval (Québec)

Top of page


Licence Creative Commons
Anglophonia – French Journal of English Linguistics est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Top of page
  • Logo Presses universitaires du Midi