Senior Lecturer Dr. Alan Reed Libert

Keynote Speach: Two Types of Polyfunctionality (and their Relevance for Endangered Languages)

Senior Lecturer Dr. Alan Reed Libert

University of Newcastle, Australia

1. Introduction

In this talk I will discuss the linguistic notion polyfunctionality and possible connections between it and endangered languages. I would like to stress that my remarks will be speculative and impressionistic, based on my years of looking at various languages, rather than being precise and statistically-based.

In theory, polyfunctionality is a simple matter: it is the property of having more than one function (or use). Many everyday objects are polyfunctional, although probably usually not by design (i.e. they were designed with only one use in mind). Take, for example, forks. Their usual use is to pick up food, but they could be used as a weapon, if one were attacked and had nothing better to fight back with. Or, for a less violent example, towels are usually used to dry oneself or something with, but one could put them in the gap between a door and the floor to keep cold or noise out.

2. Language Polyfunctionality (Sociolinguistic Polyfunctionality)

Polyfunctionality comes up in linguistics in two senses. One of these is in sociolinguistics, which is, very roughly, the study of language in societies, and I will look at this sense first, as it is the more obviously relevant one if we are concerned with endangered languages, as linguists often are. In this sense it refers to the use of a language (or dialect) in a range of contexts. For example, Langston and Peti-Stantić (2014:29) say the following:

Another important feature of a fully developed standard language is its polyfunctionality. A standard language must be able to fulfill communicative functions in all spheres of human activity, ranging from scientific texts to routine daily interactions relating to basic human needs, and it must allow its users to express complex and/or abstract ideas in a precise manner. This pertains primarily to the lexicon, which must be enlarged through the development of necessary intellectual vocabulary, but may also require the elaboration of syntactic structures.

This quotation is about the creation of a standard language (from a variety which is not a standard language), but the definition can also be related the maintenance of a language which does not aim to be a standard language, but merely to survive – perhaps it once had a fuller vocabulary and a wider range of syntactic structures, but lost them as its range of functions diminished (as they were taken over by a majority language).

If the language of a linguistic minority, e.g. Crimean Tatar, is seen as only appropriate for use at home (and this view could be held by both speakers of the majority language and by speakers of the minority language itself), younger people might decide that there is no point in learning the language or in using it with their children, and this is one way that a language can die out.

Actually, the term polyfunctionality does not come up very often in discussions of this type of situation, i.e. it is described in other terms. For example, Tamis (1990:484) states, “In Australia … however, and particularly in the country areas, MG [Modern Greek] has a restricted function in a high-contact situation”. This is not a surprise – most Australians do not speak Greek, and so one could probably not use Greek in most stores or in government offices. This situation is different from that of Crimean Tatar and of Australia aboriginal languages, because overall Greek is not in danger, at least in the near future – there are many native speakers of Greek in Greece, where they are the overwhelming majority. It would still be good to maintain Greek as a heritage language in Australia (and other countries), not least because growing up bilingual can have cognitive benefits, and might help one to get a job as an adult. On the other hand, Crimean Tatar speakers are a small minority in every country where their language is spoken, and even in Crimea (where once again one cannot generally use the language in shopping or with government officials).

There are degrees of polyfunctionality (or being restricted in function), and this largely depends on how many people speak the language. The ultimate in functional restrictedness would be a dead language, or perhaps not, because some dead languages are still used in a few contexts, and in fact some dead languages relatively recently have taken on more functions than they have had for a long time (as they were mainly restricted to being taught in universities and schools and read much more than written). There are cartoon books in Latin, Harry Potter has been translated into Ancient Greek, and Elvis Presley songs have been sung in Latin and Sumerian. Of course most such efforts are not (entirely) serious, but they do represent ideas for uses of these languages that had not been considered before.

Sadly, the vast majority of dead languages are completely dead, and not used in any way (except for linguistic analysis), and they would represent the maximum in functional restrictedness, having no functions at all. At the other end of the spectrum are languages such as Spanish, French, and especially English, which are used in many places in the world, including by many people for whom they are not native languages. However, even English is not omnifunctional, as there are some contexts in which it would not be appropriate to use it. For example, one of my French-speaking professors in Montréal said that it would be seen very badly if one Francophone spoke to another Francophone in English there (language being a political and controversial topic in Québec).

Once a language has lost some of its polyfunctionality, it is difficult to get it back. However, if speakers of endangered languages want to preserve their language, it is important to try to do so. What can be done? One concrete measure is to have signs in the language (or bilingual signs, with both the majority and the minority language). This is important, as people, whatever language they speak, see the minority language on view in public spaces, and can send various messages – this language still exists (and here is what it looks like), and it can have functions outside the home. One can see this, for example, in New Zealand, where now many signs are in Maori as well as in English. Often or usually this is merely or mainly symbolic, as speakers of the minority language will also know the majority language, and thus do not need signs in their language to find their way about.

Another measure is for a language to be used in schools – not only taught as a subject, which is important, but also to be used as the language for teaching other subjects. For example, if children in Wales are taught mathematics in Welsh, this does a couple things: 1) shows them that it is possible to use Welsh to discuss math, and 2) teaches them Welsh vocabulary in a particular area, mathematics. It may also lead to improvements in their understanding of mathematics, if they speak Welsh better than English. Of course teaching in school in a minority language often requires permission and/or assistance from governments, which are not always willing to grant this.

3. Word Polyfunctionality (Morphological/Syntactic Polyfunctionality)

Let us now turn to another type of linguistic polyfunctionality, that involving morphology and/or syntax, that is word structure and sentence structure. In fact, it, or a related and probably better known phenomenon (or more than one of them), involves the lexicon, i.e. the words in a language. We can start from the latter, as it is more likely to be familiar to non-linguists; I am speaking about homonymy and homonyms – two different words which happen to sound the same (and, with a strict definition of homonymy, are written the same as well).

Among the homonyms of English are club (‘association’) and club (‘stick for hitting people) and ear (part of the body) and ear (as in an ear of corn). I would think such pairs of words exist in many or all languages. Homonymy is to be distinguished from polysemy, which is when one word has several related meanings. We find this with the word mouth, whose basic meaning is a part of the body, but which also, by analogy, can mean the opening of a bottle, or where a river joins to the sea. I think polysemy, which we could see as extension of meaning happens in every language in the world.

Polyfunctionality, as I think of it, is somewhat different. I should note that there are a huge number of terms for it, or something like it, e.g. multifunctionality, polycategoriality, conversion, zero derivation, and the rhetorical term anthimeria. It is, roughly speaking, when words of different types have the same form, that is homonymy or polysemy, but across word types. We can think of many such cases, including in English. Consider for example, book (the noun meaning something we read) and book (the verb meaning something like ‘reserve’, e.g. “I booked two seats on the train”). Here we see one word which is able to act as two parts of speech, noun and verb. The record holder for English, as far as I know, is the word round, as shown in (1):

(1a) We sat at a round table. (adjective)

(1b) I bought a round of drinks for my friends. (noun)

(1c) He will round the corner in a moment. (verb)

(1d) The top turned round and round. (adverb)

(1e) He went around the corner. (preposition)

Not only words could be said to be polyfunctional; the same holds for some affixes. Consider for example the ablative case suffix in Turkish -den/-dan. It is used to mark the starting point of motion, but also occurs in comparative phrases. However, I will refer to this type of polyfunctionality as word polyfunctionality.

There is the question of what counts as a function when trying to determine whether an item is polyfunctional. The examples which I have given involve parts of speech (word classes), but many of the parts of speech are divided into subclasses. For example, among the verbs there are transitive verbs and intransitive verbs. However, in English many verbs can be transitive or intransitive, e.g. the verb to eat: “I ate my sandwich” (transitive), “I have already eaten” (intransitive). Should this be seen as a case of polyfunctionality? Also, in English all of the primary auxiliary verbs can also be main verbs, e.g. to be: “I am eating” (primary auxiliary), “I am tired” (main verb).

It seems to me that English has more polyfunctionality than many other languages such as Turkish, at least word class polyfunctionality, and at least if we insist on complete identity of form in different functions. This is because English is poor in inflectional affixes and nouns and verbs can easily appear without such affixes. On the other hand, in Turkish or Latin, nouns and verbs often or usually (in the case of verbs) must have at least one inflectional suffix, for case, possession, tense, person, etc., diminishing the possibility for complete homonymy between nouns and verbs.

4. Connecting the Two Kinds of Polyfunctionality

The two types of polyfunctionality which I have discusses are quite different – one is about languages, and one is about items of languages. Is there any connection between them?

Let us first set aside remarks that people might make connecting types of languages and their success on the world stage, e.g. that English has become the de facto world language due to its simplicity, or flexibility, or some such thing. For one thing, it is quite difficult to objectively measure such properties of languages – English is simple in some respects (e.g. inflection, as I have already mentioned), but complex in others: consider the large number of irregular verbs, which are so difficult for non-native learners (and even native learners – younger children often make errors with them, and adults sometimes also make such errors (prescriptively speaking)). Turkish, with its case suffixes and many verb forms, might appear more complicated than English, but Turkish does not have irregular verbs (with the possible exception of the verb meaning ‘to be’, depending on how one classifies things), so from that point of view it is simpler and easier than English.

Major reasons for the use of English all over the world are political. First, the extent of British colonization led to English being used in many areas including the US and much of Canada, various African countries, and some countries in the Caribbean. Second, the political, cultural, and commercial/industrial dominance of the United States during the second half of the 20th century have helped the spread of English. Similar reasons apply to French, Spanish, and Russian, the last of these being considered a complicated language by many people. Consider French and Romanian – they are both Romance languages, similar in structure and roughly equally easy (or difficult) to learn. French is spoken much more widely than Romanian due only to political factors, including the large number of places which France colonized.

World languages such as English and French are highly polyfunctional: they are the home language of many people, and are also used for shopping, interactions with government, etc. Romanian and Crimean Tatar are not very polyfunctional, but this is not due to any structural properties which they have or do not have. Among these properties is polyfunctionality in the second sense, that is the ability of words or other items to have more than one function (which could be seen as a part of or related to structural simplicity). In other words, at least in my view, having a high or low degree of word polyfunctionality is not directly linked to a language being highly polyfunctional or not being so. (However, it may not be clear how to measure the degree of word polyfunctionality that a language has.) French and Romanian presumably have approximately the same degree of word polyfunctionality, but they differ considerably in their range of functions.

Does this mean that there is no connection between the two types of polyfunctionality? It would appear not, but let us now bring in the subject of endangered languages. Such languages are low in language polyfunctionality, otherwise they would not be endangered, and, as I have just argued, the level of word polyfunctionality does not play a determining role in whether languages are polyfunctional (and hence endangered or not). However, consider the following: speakers of a language with a high degree of word polyfunctionality are probably aware of it at some level, e.g. English speakers are aware that many nouns can be used as verbs with no inflectional changes being made (at least in the infinitive and present tense aside from the 3rd person singular form). This is, one might say, part of what makes English English, i.e. part of its identity, clearly different from French, in which nouns generally cannot function as verbs without some ending being added.

On the other hand, this is not one of the salient features of e.g. French or Russian, because there is much less polyfunctionality (because there is more inflection).

Under pressure from a dominant majority language, various things can happen to a minority language. For one thing, it can borrow many words. Of course borrowing is not limited to endangered languages; for example, English, which is in a very powerful position in the world, has borrowed a vast number of words. (However, some of this borrowing took place when English was not in such a strong position, when England was controlled by the Norman French.) Borrowed words might be more likely to be uninflected in the borrowing language than native words, i.e. they often do not take (some of) the grammatical suffixes that are used in the native language. If this happens enough, the character of the language may change somewhat – it will have more word polyfunctionality. An example of this is described by Rochtchina (2012:73):

The category of indeclinable adjectives … described by Panov in the mid-twentieth century … is no longer perceived as a strange and exotic element of Russian morphology. In the twenty-first century, another grammatical class of analytic lexemes is being formed – the class of polyfunctional words. This class constitutes loan words predominantly borrowed from English that function as nouns and also as analytical modifiers of nouns, adjectives, adverbs and verbs. […] Through heavy borrowing from English and other languages, Russian is changing not only its vocabulary, but also its syntactic and morphological structure.

Russian is not an endangered language, but linguistic purists might be concerned about this change in Russian, as they might think that it is becoming less Russian. Imagine, however, the effect on an endangered language which, like Russian, has a fair amount of inflection: not only is the number of speakers decreasing, the language may be losing part of its character. Under the influence of uninflectable words its morphology might be simplified (i.e. there might be fewer grammatical suffixes, or those which exist might not be used as often). The morphology might also be simplified because younger people have an imperfect knowledge of the language, being only semi-native speakers. This in turn may lead to a further loss of polyfunctionality at the language level: if the minority language is perceived as being not that different from the majority language, speakers of the former might wonder why it is worth preserving.

Borrowing, even extensive borrowing, does not have to lead to the process described – a very large proportion of the vocabulary of Albanian has been borrowed, but this vocabulary has been morphologically integrated into the language, i.e. these borrowed words are inflected, as native Albanian words are. What would lead to this is borrowing with the borrowed items being uninflectable in the receiving language.

Note that being polyfunctional is not the same as being uninflectable/indeclinable. For example, the English word book is inflectable, both as a noun (it has a plural form, books) and as a verb (it has e.g. a past tense form, booked). In fact, English has few uninflectable nouns (or verbs), but, as it has many polyfunctional nouns. One might think that uninflectable words are particularly easy to borrow (since borrowing would appear to be easier if the items involved do not have any suffixes which might come along with them and complicate the process), but the many words borrowed from English (many of which are inflected) into many languages seems to argue against this. Could it be the case that what makes words more likely to be borrowed is not lack of inflection, but polyfunctionality (at the word level)? However, recall that polyfunctionality in the narrow sense which I have been using (ability to serve as more than one part of speech in exactly the same form) does depend on a language being relatively poor in inflection, so there is a link between lack of inflection, or at least paucity of inflection, in a language and word polyfunctionality.

One might think that polyfunctional words are more likely to be borrowed than non-polyfunctional ones (all other things being equal) simply because they occur more often, or at least in a wider range of contexts, and are thus more visible or salient to speakers. As far as I know, there have been no studies on this hypothesis. It may well be incorrect, but I think that it is worth studying.

One might further hypothesize that borrowed polyfunctional words are more likely than other borrowed words to be uninflected in the borrowing language. In the source language they must be able to occur without (overt) inflectional endings in at least two different functions, at least in some circumstances (as the noun book and the verb book can in English) to be polyfunctional in the narrow sense, and this feature might come along with them into the borrowing language, especially if they are polyfunctional in the borrowing language. Words which sometimes appear without inflection, including the type of word in question, might be more likely to lose all their inflection than those which always must have some inflectional affix.


Although language polyfunctionality and word polyfunctionality are not directly connected, I have have presented a type of situation in which one might have an effect on the other. A minority language which is fairly complicated in terms of morphology is in contact with, and under pressure from, a majority language which has a large number of polyfunctional words. If, as I tentatively have suggested, polyfunctional words are particularly likely to be borrowed, many such words make their way into the minority language and might be uninflected in this language, particularly if they remain polyfunctional. This, like borrowing in general, but to a greater extent, might lead to a simplification in the morphology of the minority language and to it becoming more like the majority language, having lost one of its salient properties and hence some of its identity. This in turn could cause a weakening of the polyfunctionality of the language.

This is not to suggest that borrowing is bad or should be discouraged; it is simply something that happens in languages. However, if one is concerned about the endangered languages, one might want to be aware of phenomena that increase or decrease their chances of survival.


Langston, K. and A. Peti-Stantić (2014) Language Planning and National Identity in Croatia. Palgrave Macmillan, Basingstoke, Hampshire.

Rochtchina, J. (2012) “Morphology and Lexicology Interface. Latest Russian Neologisms: The Next Step towards Analytism?”. In V. Makarova (ed.), Russian Language Studies in North America, Anthem Press, London. pp. 71-84.

Tamis, A. M. (1990) “Language Change, Language Maintenance and Ethnic Identity: The Case of Greek in Australia”. Journal of Multilingual & Multicultural Development 11.6:481-500.