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  Etymology and the European Lexicon

Reviewer: Nicholas Zair
Book Title: Etymology and the European Lexicon
Book Author: Bjarne Simmelkjaer Sandgaard Hansen Benedicte Nielsen Whitehead Thomas Olander Birgit Anette Olsen
Publisher: ISD, Distributor of Scholarly Books
Linguistic Field(s): Historical Linguistics
Language Family(ies): Indo-European
Issue Number: 29.454

Discuss this Review
Help on Posting
REVIEWS EDITOR: Helen Aristar-Dry


“Etymology and the European Lexicon” is a volume of collected papers given at the 2012 edition of the Indogermanische Gesellschaft’s regular 4-yearly general conference (with the exception of one contribution which was not given at the conference, Andrew Miles Byrd’s ‘The rules of reconstruction: making our etymologies more grounded’, 81-91). Full disclosure: I attended the conference, but did not submit an article for this proceedings. It includes 43 papers in English and German, far too many to summarise the contents of here. Despite the title, contributions are not restricted to narrow discussions of etymology, nor indeed to the European lexicon (articles cover the Anatolian, Tocharian and Indo-Iranian languages/language families, as well as Albanian, Italic, Armenian, Balto-Salvic, Germanic, Greek, and Proto-Indo-European itself - henceforth PIE). It is instead a sort of omnium-gatherum of topics which were at the forefront of Indo-Europeanists’ minds in 2012 (and somewhat afterwards: the latest reference I noted in the bibliography was to an item published in 2015).

Articles proposing etymologies of an individual word or words include: Václav Blažek, ‘On Indo-European Barley’, 53-67; Petr Kocharov, ‘The etymology of Arm. mart’ ‘possible’’, 209-17; Marek Majer, ‘The etymology of Proto-Slavic *nizъ ‘down(wards)’ and similar forms in other branches’, 267-80; Michaël Peyrot, ‘Language contact in Central Asia: on the etymology of Tocharian B yolo ‘bad’’, 327-35; Brent Vine, ‘Latin gingīva and salīva’, 479-89; Andreas Willi, ‘kakós and kalós’, 505-13. I will now list other major themes which appear in the book.

The identification and influence of substrate languages/languages in contact with (Proto-) Indo-European, including: Harald Bichlmayer, ‘Was kann man an lexikalischen und orphologischen Elementen aus dem Namenschatz der sogenannten »alteuropäischen Hydronomie« gewinnen? Ein Zwischenbericht’, 37-51; Adam Hyllested, ‘Again on pigs in ancient Europe: the Fennic connection’, 183-196; Corinna Leschber, ‘On the stratification of substratum languages’, 235-50; Biliana Mihaylova, ‘The Pre-Greek substratum revisited’, 307-17; Roland Schumann, ‘Where is the substrate in the Germanic lexicon?’, 377-84; Theo Venneman, ‘Nicht-indogermanische Spuren vorgeschichtlicher Wirtschaftsformen im toponymischen Lexikon Mitteleuropas: Seen des Salzkammerguts und Ardennen’, 443-58.

The origin and semantics of derivational morphology in PIE and its daughter languages, including: Katsiaryna Ackermann, ‘Investigating internal ways of lexicon expansion in early PIE: observations on IE roots with potential *bh extension’, 1-13; Davide Bertocci, ‘High-transitivity nasal presents between lexical etymology and morphology’, 25-36; Paul S, Cohen, ‘Reduplicated nouns in IE’, 119-34; Hannes A. Fellner & Laura Grestenberger, ‘Greek and Latin verbal governing compounds in * ā and their prehistory’, 135-149; Jay H. Jasanoff, ‘PIE *weyd- ‘notice’ and the origin of the thematic aorist’, 197-208; Norbert Oettinger, ‘Die Wechsel-Ø/n- und -i/n im Rahmen der indogermanischen Heteroklisie’, 319-26; Stefan Schumacher, ‘The development of the PIE middle in Albanian’, 385-400.

The position in the PIE language family of Tocharian and Anatolian: José Virgilio García Trabazo, ‘Zum indogermanischen und anatolischen Wortschatz der »materiellen Kultur« und seine Relevanz zur Chronologie der »nach-anatolischen« dialektalen Spaltungen’, 161-8; Melanie Malzahn, ‘The second one to branch off? The Tocharian lexicon revisited’, 281-92; H. Craig Melchert, ‘“Western affinities” of Anatolian’, 297-305.

Semantics: Bettina Bock, ‘Rekonstruktion von Semantik’, 69-80; Antje Casaretto & Carolin Schneider, ‘The relationship between etymology and semantics of local particles in the Rigveda’, 105-18; Rosemarie Lühr, ‘Basiskonzepte’, 251-66.


The book is well-produced, and includes several very high-quality diagrams and images in black and white and colour (most notably the maps on 444ff). I noticed a small number of typos throughout, and also a number of ‘non-native speaker-isms’ in the English articles; neither of which affect the comprehension of the articles nor the pleasure of reading them. The type, as usual in this series produced by Ludwig Reichert, is rather small, making the footnotes even smaller, which elderly eyes or underpowered spectacles may not appreciate.
Overall, my impression is that the quality of article in this collection is reasonably high, given its unfocussed nature and its origin in a large conference which aims to bring attendees from around the world together more than to discriminate finely among their contributions. A handful tend towards recapitulations of the authors’ previous work, or surveys of other authors’ views on a particular problem, without much new analysis. A less kind-hearted group of editors might have jettisoned these.

My own interests and abilities do not allow me to comment on every article in this wide-ranging collection. In the rest of this review I will make some comments on topics and articles that seemed to me particularly striking.

I begin with the remarkable number of articles on the identification of (Proto-Indo-European or otherwise) substrate languages mentioned above; a perennial, but not normally quite so popular, topic. This may partly be the result of the emphasis put on his version of the notional ‘Pre-Greek’ language in recent publications by Beekes (2010: xii-xlii, 2014): Mihaylova’s article explicitly intends to show that ‘Pre-Greek’ was an Indo-European language, contra Beekes’ view that it was not. These articles are split between those who attempt to identify substrate features, with varying degrees of carefulness and restraint (notably Bichlmeyer, Leschber, Mihaylova, Venneman) and Schumann, who argues that supposed examples of ‘substrate’ vocabulary in Germanic can be etymologised perfectly well as coming regularly from Indo-European elements in the normal way. This puts the finger on the general problem involved in identifying lexical items of an unknown language borrowed into a known language: what words to pick and what to compare them with. One person’s unetymologisable word, and hence substrate borrowing, is another person’s inherited Indo-European formation, either because the linguistic understanding of the development of Indo-European languages increases over time, or because of differing views on this development.

Leschber’s article demonstrates the comparison problem. After a rather unsceptical tour of ‘old linguistic strata’ in Europe and beyond, she puts forward four Balkan etyma as borrowings from a substrate. Thus, she compares Romanian mal ‘mountain, hill; shore coast’ and Albanian mal ‘hill’ to “formally and semantically similar” words in Basque, Celtic, Baltic, Slavic, north Caucasian and Dravidian languages. But the formal similarity seems to consist of a shape mVl- (where V = a vowel), with no attempt to provide a morphological analysis of the words quoted; given such a wide variety of languages to choose from and the vagueness of the semantics, it is almost surprising that she could not find more forms to quote. These problems are reminiscent of those that attend attempts to identify ‘long-range’ language relationships (on which see, for example, Ringe 1999 and Campbell & Poser 2008: 234-296). Indeed, Leschber draws on the resources of those who advocate for such relationships, while remaining vague herself about the exact status of her four Balkan words.

Mihaylova’s article is more restrained, and hence more convincing. She aims to isolate the 23 most plausible lexemes which demonstrate the existence of an Indo-European ‘Pre-Greek’ language, which show a number of consistent sound changes. Many of these are well-known and do indeed look extremely tempting, e.g. púrgos ‘tower’ next to German Burg ‘castle’ < *bherĝh , ómbros ‘storm of rain’ beside Latin imber ‘rain’. But similar problems emerge: the semantic similarity may be overly broad, e.g. ámbōn ‘crest of a hill’ beside Old High German amban ‘belly’, Latin umbo ‘boss (of a shield)’. Different etymological analyses may be available, now or in the future, e.g. ákhnē ‘chaff’ is derived from *ak-nē, with a Pre-Greek development of *k > *kh by Mihaylova; but a Greek etymology *ak-snē is equally plausible and would also give ákhnē. The very process of choosing the words is problematic: ideally, languages show regular sound-changes across the lexicon - if one is allowed to choose a very small number of words out of the mass of unetymologised Greek words and leave the rest aside, it is perhaps not surprising that one can find a number of apparent ‘sound changes’ that they have in common. Moreover, a worrying number of examples are attested only in Hesychius, a fifth century AD collection of unusual words primarily from Greek dialects, but also from a number of other languages, known from a single fifteenth century manuscript, whose evidential value is problematic.

Turning away from substrates, Byrd, in his article on ‘The rules of reconstruction’ states that “we should always ask ourselves: could the native speakers of the language in question actually have pronounced the sounds that we reconstruct for their language?” (81), and that “PIE, like all languages, possessed a grammar. A synchronic grammar” (85). It should be acknowledged that this hyper-realist approach is not one taken by all scholars in the field (compare, e.g. Clackson 2007: 16-17), for whom ‘reconstructed PIE’ “may have some features in common with the spoken IE parent language, but it is not the same as it, and it is not a real language”. For those who take a more sceptical view of the ‘reality’ of PIE, Byrd’s strictures are likely to be of less importance. Nonetheless, his proposed rules of good etymology are thought-provoking. Part of these is a reluctance to reconstruct sequences of phonemes which are reconstructed for paradigmatic/etymological reasons without being (directly or indirectly) attested in a daughter language. Of this type he argues that one can use a principle of ‘similarity’ in reconstructing sequences: thus, although not directly attested in this form, we can reconstruct *bzd- ‘fart softly’, due to its differing only in voice from *pst-, which is securely attested in *pster ‘sneeze’ and *psten ‘breast, nipple’. This is a plausible approach, but it must be used carefully; for example, it seems clear that in PIE /m/ was less sonorous than /n/ (some of the evidence for this can be found in Cooper 2013: 11-12). Consequently, we cannot assume that evidence for a sequence involving /n/ will necessarily support the reconstruction of a similar sequence involving its fellow nasal /m/.

Paolo Poccetti, ‘The Italic words for ‘moon/month’ and ‘sun’: new evidence from the Sabellian languages’, 354-63, provides the text of a recently discovered Oscan inscription and argues both that it contains the word for ‘sun’ (suleis) and that the adjective minnaris, as part of the name of a religious festival, is derived from the word for ‘moon’. It should be noted that the inscription he provides as a parallel for his analysis is Cp 30 in Rix (2002), rather than Cp 29 as printed (354). Although the argumentation he provides for the meaning ‘sun’ is strong, I am not entirely convinced: the context is not entirely certain, and such a form causes phonological problems. It looks like an exact counterpart of Latin sol ‘sun’, which is best reconstructed as *sōwol, although *sāwōl is also conceivable. Neither of these would be expected to give Oscan suleis, since intervocalic *w is not lost in Oscan. Poccetti’s suggestion that both forms come from *sh2wōl > *swōl requires a starting form which does not fit with the known evidence for the paradigm of this word in PIE (Zair 2010/11 [2012]: 210-11), and does not address the problem that *sw- is normally retained in Oscan, as shown by e.g. svai ‘if’. An alternative possibility that suleis goes back to *sh2ul- > sūl- is more plausible (it would then be identical to Old Irish súil ‘eye’, which is generally assumed to have undergone a semantic shift from ‘sun’ via a metaphor of the sun being the ‘eye of the sky’). But long *ū usually gives -ī- in Oscan, as in tiium ‘you’ < *tū-om.

For minnaris, Poccetti reconstructs *mēnā-, with the ‘littera rule’ whereby a sequence of long vowel followed by a single sonorant becomes a short vowel followed by a geminate sonorant. The etymological connection with ‘moon’ is plausible, but not the supposed developments. There is no other good evidence that the littera rule took place in Oscan; while in Latin, where it definitely did occur, it took place only in the sequences ‘high vowel + voiceless stop’, ‘/a/ + sonorant’, and ‘front vowel + /l/’ (Sen 2015: 42-78). Rather, minna- is probably derived from *mēns-(V)n-, with the n-enlargement seen in the related languages Umbrian menzne, Marsian mesene. In general, Poccetti is a little too trusting in the consistency of the spelling of Oscan: he states that the spelling variation of the first vowel in fisia-, fiisia- and fiísia- ‘festival’ may be due to synchronic and diachronic variation in the language, and that the use of a single <i> in minnaris reflects a genuine short /i/, since <ii> or <ií> are used for a long vowel. But, as is clear from looking at the corpus of Oscan, in the Oscan alphabet there was simply free variation between use of single or double letters to write long vowels (and geminate consonants), which can often be found even within individual inscriptions. A further case where the spelling is important is on p.354, where fusent is accurately identified as a 3rd plural future indicative in the main text, but footnote 3, apparently referring to fusent, states “[f]or textual and syntactic reasons, the imperfect subjunctive is more likely than a future indicative” (the footnote does not seem to follow on well from its position in the main text; I suspect it is left over from an earlier draft). It is certain that fusent must be future indicative: the imperfect subjunctive would be written fusins.

Elena Triantafillis, ‘-d- verbal bases (claudo, mando, plaudo…) between Latin and Indo-European: an etymological analysis’, 415-25, argues that a number of Latin verbs whose roots seem to end in a -d- which is lacking in the related verbs in other Indo-European languages should be considered to be originally compounds involving the verb do, dare ‘give’ (from PIE *deh3-). This is already the standard analysis of uendo ‘I sell’, which transparently consists of uenum ‘for sale’ + do, as further demonstrated by its perfect uendidi, which maintains the reduplicated formation seen in dedi ‘gave’ (see also the compounds credo, credidi ‘believe’, condo, condidi ‘build’ etc.). For the other verbs considered here, however, the suggestion is a new one. Triantafillis also raises the possibility that at least some of these forms really reflect the root *dheh1- ‘put’, which has similar ‘light’ semantics to ‘give’ and would give the same phonological result as do < *deh3- in the middle of a word in Latin. Strangely, Triantafillis does not mention the counterpart of mando ‘order’, in the related language Oscan, whose perfect is manafum. Since *dh gives -f- in Oscan while *d remains as -d-, this demonstrates that we have the ‘put’ root rather than the ‘give’ root, for this verb at least. Overall, the idea that the Latin verbs in -d- reflect old compounds involving do and *dheh1- is plausible, since we know that some of the verbs already come from such an origin. Presumably the variation in perfect formations (uendo, uendidi but e.g. mando, mandaui and pendo, pependi ‘hang’) is due to the creation of the constructions at different times in the history of Latin, with older forms becoming more opaque and hence being restructured to fit into the productive types of perfect formation at the time.


Beekes, Robert S. P. (2010). Etymological Dictionary of Greek. With the assistance of Lucien van Beek. Leiden & Boston: Brill

Beekes, Robert S. P. (2014). Pre-Greek: Phonology, Morphology, Lexicon. Edited by Stefan Norbruis. Leiden & Boston: Brill

Campbell, Lyle & William J. Poser (2008). Language Classification. History and Method. Cambridge: Cambridge University Press

Clackson, James (2007). Indo-European Linguistics. An Introduction. Cambridge: Cambridge University Press

Cooper, Adam I. (2013). The typology of PIE syllabic sonorants. Indo-European Linguistics 1, 3-67

Ringe, Don (1999). How hard is it to match CVC roots? Transactions of the Philological Society 97, 213-44

Rix, Helmut (2002). Sabellische Texte. Die Texte des Oskischen, Umbrischen und Südpikenischen. Heidelberg: Universitätsverlag C. Winter

Sen, Ranjan (2015). Syllable and Segment in Latin. Oxford: Oxford University Press

Zair, Nicholas (2010/2011 [2012]). British *-āw- and *-āg-, and the Celtic words for ‘sun’. Die Sprache 49, 194-208
Nicholas Zair is Lecturer in Classics (Classical Linguistics and Comparative Philology) at Cambridge University.

Format: Hardback
ISBN-13: 9783954902026
Pages: 552
Prices: U.S. $ 177.00