Review Society

Endangered Languages

About: Nicholas Evans, Ces mots qui meurent. Les langues menacées et ce qu’elles ont à nous dire, La Découverte, 2012, 396 p., 28,5€.

by Denis Bertet , 27 April 2015
translated by Victoria Lazar Graham

Translated with the support of The Florence Gould Foundation

The translation into French of Nicholas Evans’ book Dying Words. Endangered Languages and What They Have to Tell Us is a timely reminder for Francophone readers of the magnitude and, particularly, the severity of a phenomenon on which very little has been written in the French language and of which there is still far too little awareness: the mass extinction of languages currently underway.

Reviewed : Nicholas Evans, Ces mots qui meurent. Les langues menacées et ce qu’elles ont à nous dire (Dying Words. Endangered Languages and What They Have to Tell Us, Wiley-Blackwell, 2010), translated Marc Saint-Upéry, Paris, La Découverte, 2012, 396 p., 28,50 €.

Understanding linguistic diversity

Current estimates suggest that the world loses around 25 languages every year. If such loss rates continue, between a third and one half of the 5-7 thousand languages spoken today (according to the current definition of what constitutes a language) will have disappeared by 2100, to say nothing of the possibility, and indeed the likelihood, that this phenomenon will go on to accelerate over the course of the 21st century as previously relatively isolated regions gradually integrate into their nation states. The bleakest prognosis, of which American linguist Michael Krauss [1] is a notable proponent, goes so far as to predict a 90% loss in languages by the end of the century.

Some would have us believe that this is no bad thing. They are not, of course, celebrating the loss of linguistic diversity, but rather what they assume to be its natural consequences: cultural homogenization, a stronger sense of national unity and improved communication between peoples as a direct result of linguistic unification; in short, peace and economic growth. Things are never quite so simple, however, and these oft-repeated arguments are undermined by several rather glaring examples; as British linguist David Crystal rightly points out, the remarkable linguistic homogeneity of Rwanda, for example, in which the overwhelming majority of the population use dialects largely comprehensible to speakers of the country’s official language, Kinyarwanda, did little to prevent what was one of, if not the, worst genocidal civil wars ever to occur in what was nevertheless a predominantly multilingual Africa. Likewise, Switzerland’s multilingualism has in no way hindered its economic development. Thus, while shared dialects can forge links between communities, it is by no means certain, and depends on a multitude of other factors. It is entirely possible for an official state language shared by all to exist alongside regional dialects without necessarily posing any threat to their existence.

Others, scarcely more balanced in their outlook, have been more alarmist, likening language death to the loss of a great treasure, arguing that this would inevitably result in terrible cultural impoverishment. Some even go so far as comparing language extinction to the plight of endangered species, which, if they were to disappear, would have disastrous consequences for the planet as a whole, as we are well aware. The time has come, they argue, for us to act urgently and protect our linguistic diversity. No need, it seems, to overly concern ourselves with the opinions of those most directly affected, namely the communities who actually speak or who once spoke those endangered languages. What stands out as much in this discourse as it does in the opposing one put forward by supporters of linguistic unification, and what is also disturbing, since there is ultimately some truth to these arguments, are the false assumptions, the mental shortcuts, the metaphors and general ambiguity upon which the arguments are founded. Their vaguely prophetic warnings of language extinction have clearly failed to move both political and social actors, as well as the wider general public. It could even be said that they appear to harbour an ideology of diversity at any cost, whatever form that diversity might resemble, which, despite challenging the equally problematic ideology of integration at any cost, is still no less ideological in nature.

What both sides, as well as the general public, sorely lack in this important but still largely neglected area of debate, is above all a precise and well-informed understanding of the actual issues at stake. What are the implications of reduced global linguistic diversity, and who stands to win or lose? The role of linguists is a crucial one, for it is they who are best placed to provide an overview of the debate and to contribute and share information as objectively as possible so that the public and other diverse actors may develop informed opinions on the matter.

What is lost when a language dies

In Dying words, Nicholas Evans, Australian field linguist and specialist in the Aboriginal languages of Australia, chooses to focus on what we stand to lose from language death (others have written about what may be gained). It was through his work on Australia, and the fact that Australia currently suffers from one of the highest rates of ‘linguistic erosion’ in the world, that he first learned of the issue of endangered languages. It is unfortunately rare to find a text in which the potential losses to humanity are discussed in such a clear and measured manner, using language that is both rigorous and precise enough to avoid vague assumptions and the use of poorly-applied comparisons and imagery, yet which retains a very fluent, human quality, accessible to all curious minds. This is one of those very rare texts which discusses, in a way that holds meaning for all, the concept of ‘endangered human diversity’, a concept invoked far too frequently in simplistic arguments, and which as a result, has had the unintended effect of further fuelling the scepticism of critics who view these arguments as being almost mercantile in nature. Whatever they decide, it is through honest, objective information, such as that provided by Evans, that the public and other relevant actors are most likely to engage with the debate.

So, what exactly is at stake? Evans looks in far greater detail at what the speakers of dominant languages stand to lose than he does at the implications for speakers of endangered languages, undoubtedly not only because the former are more likely to read his book but also because it is they who have the power to ensure that the latter are afforded a certain degree of freedom to decide their futures, both regarding legislation but also, crucially, in practical terms. The loss for dominant language communities would essentially only be intellectual in nature.

Naturally, it is the linguistic sciences which stand to gain the most from linguistic diversity (discussed in Part II). The exploration of the exotic (from a Western perspective, of course) peculiarities of Navajo, a well-documented Native American language spoken primarily in Arizona and New Mexico, runs like a common thread throughout the author’s career, and he shows how the study of diverse languages is crucial to our understanding of linguistic possibilities, as much at the practical level of how a language functions (through sounds and, in the case of sign language, gestures) as at the level of signifier (morphosyntax) and signified (semantics). It is not unheard of for field linguists and typologists (specialists in typology, the branch of linguistics which aims to measure diversity within the universal phenomenon of language) to occasionally identify a language which exhibits linguistic characteristics considered highly unlikely or even impossible. Discoveries such as these undeniably enrich our understanding of human language. The author points to the example of Kayardild (p. 6-7), an Australian Aboriginal language, on which he published a grammar book in 1995, and which very convincingly challenges the statements made in 1990 by psycholinguists Paul Bloom and Steven Pinker that no language used nominal affixes to express time. [2] In fact, Kayardild, and a few other languages to a certain extent, express the tenses (present, past, future etc.) not only through verbs, as we usually do, but also through all accompanying nominals (subject, complement etc.) Were it not for the discovery of these rare languages, in which time is expressed so uniquely, linguists would have continued to labour under their former misapprehensions.

The study of diverse languages, which offers increasing precision and temporal depth the greater the number of languages studied, also teaches us about the history of those languages and of the people who speak or who once spoke them, as well as, very often, their prehistory, since the vast majority of world languages have only very recently been documented in writing (Part III). It is the study of the Yeniseian languages, a language family once spoken in the Yenisei River region of Siberia, and from which only one language, Ket, survives, with less than 500 speakers, which, coupled with research on the Na-Dene, a Native-American language family, lead researchers to suggest a historic link between the two language families of Asia and America (p. 189-195). The first waves of migration to America are believed to have occurred no earlier than 11,000 BC. Assuming, as is likely, that other groups migrated after this date, the latest groups to migrate should logically have done so recently enough for these links between the Native American language families (spoken by the first groups to travel to America) and the Asian language families (spoken by the first groups which remained in Asia) to be possible. It was not until the 2000s, when a genetic cluster between the Yeniseian and the Na-Dene families was proposed on the basis of some admittedly shaky similarities, that any serious attempt was made to link the two families. If this hypothesis can be confirmed, it will have been through the study of these languages, almost all of them endangered, that we will finally have been able to establish with historical certainty a shared history between certain Native American and Asian populations. Similarly, it is primarily through the study of rare and endangered languages, namely the Mixe-Zoque languages from the south of Mexico, that we have been able to decipher the writings of the Olmec, most likely Mesomerica’s most ancient civilisation (p. 223-233).

Linguistic diversity also has much to teach us on the subject of the human condition and our creative abilities (Part III). While language is by no means a direct reflection of cognition, the particular, sometimes extremely rare, characteristics of the world’s many languages certainly, in one way or another, influence the thought processes of those who speak them. Unless it is the way one thinks which influences language. Or perhaps both processes occur. Linguistic diversity may also offer a somewhat indirect insight into the scope of human cognition. Would speakers of Western languages, who rarely refer to the points on a compass for any purpose other than to determine geographical location, and who usually only apply relative notions of left and right, not think differently to speakers of Kayardild, who define virtually every situation and movement according to its compass direction, and who thus seem constantly aware of their position on the compass? (p. 248-257). While these are sensitive questions, it is clearly of benefit to the cognitive sciences for this type of discussion to emerge from the study of linguistic diversity.

The author’s simple, clear argumentation, his rich and wisely-chosen examples and the useful illustrations will appeal to readers unfamiliar with the topic, while more informed readers, who may already have specialist knowledge of the subject, will also appreciate this pleasant journey through the linguistic regions of Australia, the Americas and Papua, which, despite featuring frequently in Ango-Saxon research, have been woefully neglected in continental Europe. This latter readership will also benefit from being reminded of the importance of the many sign languages within the greater context of linguistic diversity, particularly since current challenges in documentation (which transcription methods would be most effective? Or should we use video recordings?) leaves them especially vulnerable to disappearing without a trace (p. 73-77).

What next?

Now that we know what is at stake, how are we to proceed? In terms of how we should react to language loss, there are several reasons why the author chooses not to address those communities most directly affected by the phenomenon. Some of these are explained in the section entitled ‘Further Reading’ which can be found in the original edition but which was unfortunately omitted from the French translation: ‘The difficult challenge of what small communities can do to maintain their languages is a topic I decided not to tackle in this book, partly because there were already so many other topics I wanted to cover, but also because it is such an uphill battle, with so few positive achievements, and as much at the mercy of political and economic factors as of purely linguistic ones.’ [3] As the American phonetician Peter Ladefoged famously pointed out to his colleague Kenneth Hale [4], why should these communities not be entitled to make choices regarding their particular circumstances, even if, much to linguists’ displeasure, they then choose to abandon their language in favour of a dominant language that they perceive, rightly or wrongly, to be more useful? Nor does the author address the general public and the relevant social and political actors, for, there again, it is up to them to develop their own opinions, however uninformed they might be.

Instead, he turns primarily to his fellow linguists, and more or less implicitly and with great lucidity explains that while their role remains first and foremost to provide objective information to the communities affected, the public and the relevant actors on how to respond to language loss, it also includes the responsibility to record to the best of their ability what remains of the languages before they disappear (Part V). It is that which should form the focus of their activities. The author points out several shortcomings in the current study of linguistics, of which the worst is the failure to encourage, and even to actively discourage, at least at university level, the study of unknown or little-known languages:

Since the ascent to dominance of Chomskyan generative linguistics in the 1960s, the focus in North America and in many countries that followed its trends has been on theoretical modelling of fragments of well-known languages, rather than on new empirical work. Indeed, it is currently the case that in most US universities a reference grammar of a little-described language is not a permissible doctoral topic, despite the fact that it is about the most demanding intellectual task a linguist can engage in. (p. 329-330).

While this tendency is certainly less visible among French academics, field linguistics is still treated as a subject at the margins, and the practical training for this particular area of linguistics even more so. While it would be wholly unfair to blame linguists for language death, they may, on the contrary, be held fully accountable for failing to adequately document where possible those languages which could disappear without record. It has also become clear that the up-and-coming field of ‘documentary linguistics’, which aims to record as many speech samples of as many varieties of language as possible without being able to actually analyse or to even transcribe them (in the case of oral samples), should be viewed as no more than a last resort, since too often these samples are practically useless. Ideally, each language should have its grammar, a dictionary and a properly transcribed encyclopædia written up for it; these are the types of ‘traces’ linguists should be responsible for maintaining (p. 331-336). Another task for today’s linguistics would be to better promote the professional training of linguists who are native speakers of the languages they are researching, if only because, as we are increasingly aware, the linguistic sciences gain immensely from contact with both internal and external perspectives on languages (p.324-329).

In Dying Words, Nicholas Evans very successfully fulfills the double task he sets himself in his prologue; to write a book that is ‘about everything that is lost when we bury a language’s last living witness, and about what we can do to bring out as much of their knowledge as possible into a durable form that can be passed on to future generations,’ in a way that speaks in equal measure to ‘linguists, the communities themselves and the lay public’(p. 9-10).

To quote this article :

Denis Bertet, « Endangered Languages », Books and Ideas , 27 April 2015. ISSN : 2105-3030. URL :

Nota Bene:

If you want to discuss this essay further, you can send a proposal to the editorial team (redaction at We will get back to you as soon as possible.

by Denis Bertet , 27 April 2015


[1Kenneth Hale, Michael Krauss, Lucille J. Watahomigie, Akira Y. Yamamoto, Colette Craig, LaVerne Masayesva Jeanne, Nora C. England, ’Endangered Languages’, Language, vol. 68, n°1, 1992, p. 4-10.

[2Steven Pinker, Paul Bloom, ‘Natural Language and Natural Selection’, Behavioral and Brain Sciences, vol. 13, 1990, p. 707-726.

[3p. xix of the original edition.

[4Peter Ladefoged, ’Another View of Endangered Languages’, Language, vol. 68, n°4, 1992, p. 809-811.