Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  Handbook of Comparative and Historical Indo-European Linguistics

Reviewer: Geoffrey Sampson
Book Title: Handbook of Comparative and Historical Indo-European Linguistics
Book Author: Jared Klein Brian Joseph Matthias Fritz Mark Wenthe
Publisher: De Gruyter Mouton
Linguistic Field(s): Historical Linguistics
Issue Number: 29.2388

Discuss this Review
Help on Posting

Until now, the only comprehensive reference work on the development of the various Indo-European languages out of Proto-Indo-European has been Karl Brugmann and Berthold Delbrück’s “Grundriß der vergleichenden Grammatik der indogermanischen Sprachen”, first published in eight volumes from 1886 to 1900. In 1968 what was labelled as the second volume of a new “Indogermanische Grammatik” was published, projected to run to many volumes, but fifty years later that enterprise has proceeded only partway to completion and Jean-Claude Muller (p. 215 of the book under review) sees it as “embarked on a downhill path”. Yet the need is there: knowledge about the history of the IE language family has obviously expanded greatly since the nineteenth century. (As one index of that, two of the twelve main IE subfamilies recognized by Klein et al., Anatolian and Tocharian, were unknown to Brugmann and Delbrück.) Now the book under review (although not identified on title page or cover as “Volume 1”) presents itself as the first volume of what we are promised will be a three-volume replacement for Brugmann and Delbrück. It contains a Table of Contents showing the intended structure of the whole set of volumes.

(This volume has no index, and the Table of Contents does not suggest that there will be one in a later volume. If not, for a book like this that would be a very regrettable omission. The “Grundriß” had a whole volume of indexes.)

The new undertaking has been approached in a very different style from Brugmann and Delbrück’s. Rather than being written by two or a few co-authors, the 21 (existing and planned) chapters are each divided into sections, often just a few pages long, contributed by different scholars; according to the Preface there will be a total of 120 contributors apart from the co-editors. (No information is given about the contributors; unusually, their names are shown at the end of their respective sections with the name not of the institution where they work but of the town or village where they live.) Many sections were drafted ten years ago (with minimal opportunity for their authors to revise them since, so literature references are not always as up to date as they ideally might be).

The “architecture” of the new work contrasts with that of the “Grundriß”, too. Brugmann and Delbrück began from Proto-Indo-European and worked forward, showing how PIE split into its main “dialects”, such as Greek, Germanic, Celtic, and so on, and how these evolved in turn into mediaeval and present-day languages. The work under review, on the other hand, is what Jared Klein (p. vi) calls “inductive”, meaning that it begins with the earliest concretely attested representatives of the main subfamilies, and from these works forward to the modern languages and backward towards PIE.

This first volume comprises the following chapters (I abbreviate some of the titles):

I General and methodological issues
II Application of the comparative method to language-families other than IE
III The history of IE linguistics

And chapters IV to VII cover four of the twelve subfamilies, namely Anatolian, Indic, Iranian, and Greek. Within these subfamily chapters, the sections follow a common pattern: where “X” is the respective subfamily, the sections are:

* the documentation of X
* the phonology of X
* the morphology of X
* the syntax of X
* the lexicon of X
* the dialectology of X
* the evolution of X

– except that Anatolian, having gone extinct, has no “evolution of” section. The two further volumes will cover the other subfamilies, together with “Languages of fragmentary attestation”, such as Phrygian and Messapic, and chapters on Proto-Indo-European and on possible remote relationships between PIE and sister language-families.

Among the opening chapters, Chapter I includes sections on the nature of relationships between languages, the evidence available for reconstructing early states of IE languages, the various writing systems used for IE languages, and on what is known or surmised about the culture and homeland of the speakers of Proto-Indo-European. Chapter III recounts the history of the initial realization that the IE languages form a genetically-related family (Pierre Swiggers, p. 159, sees Sir William Jones’s famous lecture of 1786 about Sanskrit as less crucial here than it is usually described as being), and the subsequent development of historical linguistics as a discipline.

Chapter II is less straightforward to summarize. Its six sections are titled “The comparative method in X linguistics”, where X ranges over the non-IE (sub)families and areas “Semitic”, “Uralic”, “Caucasian”, “African”, “Austronesian”, and “Australian”. What is not straightforward is to grasp how these sections are intended to contribute to the book as a whole. The six topics do not cover the whole world, and might be said to overlap. (Semitic is a subfamily of the Afro-Asiatic family, so not entirely separate from “African”.) These sections are not obvious components for a book on the IE language-family, and we are not told why the editors included them.

The answer may be different for different sections. Semitic and Uralic are perhaps included because it is likely or certain that they had early contacts with IE languages, and some scholars have even believed that they could be remotely related to IE. That is obviously not so for Australian languages (unless one were to accept the extremely controversial idea that all languages of the world share an ultimate common origin), but the “African” and “Australian” sections are interesting because they show that problems facing those who try to reconstruct protolanguages for other families can be rather unlike the problems which commonly arise for IE.

Thus Rainer Vossen points out that the principle of regular sound correspondence can be difficult to apply to African languages because “it is often not so easy to find even two examples of a given regular correspondence”, one reason being that “in languages with inflated sound inventories some sounds are (much) rarer than others. In Central Khoisan, for instance, consonants such as ‘tx’ or ‘dz’ occur in just a handful of words”. Paul Black, writing about Australian linguistics, questions how far the comparative method is applicable at all “in a continent with hundreds of speech varieties that seldom had as many as a thousand speakers each”. Among other specific problems, some Australian languages apparently seem to have undergone, independently of one another, particular types of sound change (e.g. loss of initial consonants or initial syllables) that are unusual on a world scale. Often, even core lexical items have been borrowed so widely between languages that it is hard to distinguish language contact from common ancestry. And semantic relationships can be mysterious: in one Australian language-family, five separate words in the protolanguage each developed into words meaning “one” in some descendant languages and “raw” in others, though the connexion between these two concepts is opaque.

Historical linguistics has been so heavily moulded by research on IE languages that it is healthy to be shown that assumptions which seem to work well for these languages do not necessarily hold up on a world scale. Among scholars of IE, the Neogrammarian hypothesis that sound-laws apply across the board to all words meeting their input conditions has been accepted very widely; the few linguists, such as Hugo Schuchardt, who have questioned it are seen as something like heretics. Yet researchers on the history of Chinese seem mostly to have found it natural or unavoidable to see sound changes as diffusing gradually through the lexicon, and sometimes running out of steam before they have applied to all the words to which they might apply. Is this a real difference between language-families or merely a difference of current consensus between largely-separate scholarly communities; has the Neogrammarian hypothesis perhaps blinded IE linguists to data which could amount to evidence against regular sound-laws? The inclusion of Chapter II (though, as it happens, Chinese is not one of the languages covered there) encourages such questions to be asked and, one hopes, eventually answered.


De Gruyter Mouton have done the field a considerable service by taking on this large-scale publishing project. A replacement for Brugmann and Delbrück was long overdue, particularly now that historical linguistics has been reinvigorated after a relatively fallow period around fifty years ago. There seems little doubt that this set of volumes will become a standard starting-point for linguists wanting to check details about areas of IE history where they are not themselves expert.

That said, of course there are respects in which the book arguably falls short of perfection. I wonder how wise it was to divide the work between more than a hundred contributors. I can well believe that no two scholars now would be equal to covering the entire field, as Brugmann and Delbrück did. But one might think that the various aspects of an individual subfamily could all have been covered by a single author. If the number of contributors had been kept within bounds, avoiding the blizzard of correspondence which must have confronted these editors, they would have had more time and energy to spare for resolving contradictions and eliminating repetitions between contributions, cajoling authors into interpreting their briefs in a consistent way, laying down what level of prior knowledge by readers should be assumed, and so forth. (As it is, Jared Klein’s editorial preface gives several hints at the frustrations which beset the enterprise.)

As an example of contradiction between contributions, on p. 81 Stefan Zimmer, on “The culture of the speakers of PIE”, gives ca 3000 BC as a likely date for when PIE was spoken (which I believe is in line with the general consensus), but on p. 85 Torsten Gaitzsch and Johann Tischler, on “The homeland of the speakers of PIE”, suggest “somewhere between 10,000 and 5,000 BC”. At both places the authors stress the uncertainty surrounding their dating, and I do not suggest that they should have been asked to bring their estimates into closer agreement: arguably it is a virtue for the book to illustrate the diversity of scholarly opinion. But for readers it would be less confusing (and potentially misleading) if Gaitzsch and Tischler had mentioned that others prefer a lower date (and vice versa for Zimmer).

The same two sections illustrate the different interpretations placed by contributors on the tasks set by the editors, who evidently had a consistent plan for the book in their own minds which has sometimes been distorted by individual contributors. In the “PIE homeland” section Gaitzsch and Tischler identify the various areas, Ukraine–South Russia and others, which at different times have been proposed as the IE “Urheimat”, backing this material up with references to the scholars who made the respective proposals and the types of evidence they used. On the other hand Zimmer, in his “PIE culture” section, focuses almost exclusively on detailed bibliographical references to the literature on that issue, with scarcely any information about the actual content of the respective publications. Either approach is reasonable, but I feel sure that the editors will have hoped that both contributions would adopt a similar approach.

Again, we have seen that each chapter on one of the IE subfamilies begins with a section on “The documentation of [that subfamily]”, and for most of the subfamilies covered in this volume the topic is treated by surveying the documents in which the earliest variety(ies) of the subfamily is/are attested, together with references to the modern scholarly literature on it. Gregory Nagy’s “The documentation of Greek”, though, is quite different: much of it might best be described as a philosophical disquisition on the meaning of the term “documentation”. (It is true that early Greek is so abundantly attested, and the scholarly literature on it so massive, that a section adopting the majority approach might have been impossible to keep within bounds.)

There are issues about the level at which different contributions are pitched. Occasionally, contributions spell things out more than seems necessary. Bernard Jacquinod, on p. 683 of “The syntax of Greek”, explains in some detail what it means to be an inflecting language – a reader unfamiliar with that concept would be unlikely to get much out of this book. More problematic are cases where contributions assume knowledge of technicalities which need explanation, as when Masato Kobayashi on “The phonology of Indic” refers on p. 328 to “plutic lengthening” (a Sanskrit-specific term) and uses a contrast between plain and boldface phonetic symbols which I do not understand, or José Luis García Ramón on p. 661 of “The morphology of Greek” makes a glancing reference to “Osthoff’s Law” (a vowel-shortening sound change which applied in the transition from PIE to Greek). Contributors sometimes use words like “consequently”, “therefore”, in cases where the logical relationship between antecedent and consequent depends on language-specific facts that are not made explicit. And Vit Bubenik on p. 647 of “The phonology of Greek” loses me when he quotes an Elean spelling <boikiar> for Classical /oikíās/ as evidence for the dating of the sound-change by which Classical /b/ has become Modern Greek [v]. How is <b> for Classical zero relevant to that sound-change? Likewise I am baffled by Pierre Swiggers’s remark on p. 140 that comparative linguistics requires a concept of a linguistic domain “over which the relationships (Σ R) extend”: these algebraic symbols are never mentioned again, and they have no standard linguistic use that I know of.

These are problems which might have been eliminated by more activist editing. Most often, though, when passages in this volume are hard to follow because of unstated background knowledge, this seems to be a consequence of the editors’ deliberate choice of an “inductive” architecture. In theory it sounds sensible to begin from known languages and work towards the hypothetical PIE. The trouble is, it is difficult to tell any story backwards. Contributors obviously know how features of the early attested languages relate to properties of PIE as currently reconstructed, and they frequently refer to PIE properties in order to explain the former features – it would be artificial not to. But that means that many passages in this first volume, for instance references to “ablaut”, “zero grade”, “full grade”, depend on material that will appear only in the third volume. Someone with access to the full set of volumes will be able to refer forward, and I could usually see what was meant based on my very sketchy awareness of PIE supplemented with other published accounts of it on my own shelves. But it feels rather unnatural when even a quite general section in Chapter I, Don Ringe’s section on “IE dialectology”, includes a casual reference to “simple thematic presents” (p. 66) which within the framework of this set of volumes will only be given a meaning many hundreds of pages later in Chapter XX.

(Briefly, “thematic” v. “athematic” refer to two inflexion patterns, inherited from PIE by some of the successor languages. In PIE roots the consonants were fixed but most vowels varied depending on word-structure, and in some circumstances a root would have no vowel; “ablaut” is this vowel variation, traces of which survive in modern IE languages including English, and “zero grade” is the vowel-less form of the root.)

I found the issue about “telling the story backwards” specially bothersome in connexion with Greek accent placement, which has long mystified me and is handled in Vit Bubenik’s section already cited. Of the languages for which I have at least a rudimentary speaking ability, Modern Greek is the only one where I rarely know which syllable to stress – and, surely not coincidentally, it is the only one whose usual orthography marks stress on every word. Bubenik discusses various traditional rules relating to accent placement in Classical Attic Greek, but those rules only really amount to specifying syllables which cannot take the accent (because they are too far from the end of the word): they do not say which syllable does take it. From what Bubenik says, it sounds likely that the apparently patternless Greek situation resulted from sound changes applied to a PIE situation which was perhaps more transparent, but the structure of the book means that we do not learn about this.

Inevitably I found myself querying or disagreeing with some propositions asserted by various contributors. Jay Jasanoff in “The impact of Hittite and Tocharian”, pp. 233–4, claims that the traditional picture of the breakup of PIE postulated a “big bang” in which the ancestors of the main subfamilies all emerged at once. Yet already in the 1860s August Schleicher published a family tree showing the nine subfamilies he knew about descending from a unitary PIE through exclusively binary branchings at different periods (the initial split being between “slawodeutsch” and “ariograecoitalokeltisch”). I did not understand Gaitzsch and Tischler’s claim (p. 88), in connexion with a possible route of IE migration, that “the share of Neolithic DNA” decreases across Europe from SE to NW. Do human beings everywhere not have about the same number of ancestors who lived during the Neolithic age? Hans Hock (“Language contact and IE linguistics”, p. 7) claims that when originally different languages converge, “it is structural features that come to be more similar, while the lexicon tends to remain distinct”. I do not know about convergence between European languages, but I do not think that what Hock says can be a general rule. For instance, it was rather the other way round when Vietnamese became a language which feels now like a somewhat idiosyncratic dialect of Chinese, despite being genetically unrelated to it. (Unfortunately Hock’s citation for this point, “Gumperz and Wilson 1971”, is omitted from his reference list.) Caley Smith (p. 434 in “The dialectology of Indic”) seems to contradict himself about whether Standard Punjabi has a two- or three-tone system. Peter Daniels in his “writing systems” section, p. 42, has missed the fact that the distinctive script previously used for Irish was given up in favour of roman in the 1960s.

I was puzzled by a comment Stefan Zimmer makes (p. 76) about the name “Indo-European”. Zimmer deplores the fact that it has displaced the alternative name “Indo-Germanic”; he sees the former as “partly misleading and rather unhappy, and not at all ‘politically correct’ ”. But the terms seem equally logical as descriptions of a group of languages spoken over much of the territory between India and Iceland (without Iceland, perhaps “Indo-Celtic” would be more logical than “Indo-Germanic”). I am no expert on political correctness, but I have always supposed that “Indo-Germanic” fell out of favour as a reaction against the grim political record of Germany back in the early twentieth century. However, for scholarly purposes all that matters is that we know what we mean by the terms we use.

All contributions are in English, and the language is readable throughout, considering that many contributors are not native speakers. (If occasional phrases of foreigner-English had been edited out, we would have been deprived of Masato Kobayashi’s delightful remark that, in Sanskrit, “A vowel is taught to belong to the same syllable as its onset” – meaning, doubtless, that Sanskrit grammarians teach this principle of syllable division.)

I noticed a scattering of minor misprints or oversights: “Undersogelse” for “Undersøgelse” in Rasmus Rask’s book title (p. 6); “a” missing from “Monatsschrift” (p. 183); on p. 220 “familiar” should surely be “unfamiliar”; the Greek particle ἵνα is transliterated in two different ways on p. 727. Peter Daniels’s remark, p. 46, about “syllabic characters (for the 3 stops Ø or plus each of the 5 vowels)” seems garbled. But the fewness of errors like these just serves to underline the fact that in general the book has been very well produced.

I cannot close this review without a comment on the price of the book. German publishers commonly charge eyewatering prices by British or American standards, and I cannot criticize De Gruyter for making this book so expensive: it may well be that had the project been proposed to Anglophone publishers, they would have declined to take it on, doubting that it could be made to pay at a price which the English-speaking academic community would find acceptable. The fact remains that a book like this yields most value if relevant scholars can keep a copy by them, make notes in its margin, and so forth. Regrettably, Klein et al. will be consulted only in libraries.
Geoffrey Sampson graduated in Oriental Studies at Cambridge University in 1965, and studied Linguistics and Computer Science as a graduate student at Yale University before teaching at the universities of Oxford, LSE, Lancaster, Leeds, and Sussex. After retiring from his Computing chair at Sussex he spent some years as a research fellow in Linguistics at the University of South Africa. Sampson has published in most areas of Linguistics and on a number of other subjects. His most recent book is ''The Linguistics Delusion'' (Equinox, 2017).

Format: Hardback
ISBN-13: 9783110186147
Pages: 732
Prices: U.S. $ 378.99