A great deal is known about the Indo-Europeans; a tale of a hero who *ég hent óg him ‘slew the serpent’ was central to their mythology; their poets sang songs of heroic deeds to gain their patrons and themselves *méĝh₂ $\hat{k}$ lé os ‘great fame’; a royal sacrifice of theirs involved copulation with a horse.Footnote ¹ These feats of knowledge bear witness to the labors of generations of linguistics and philologists working directly with traditional literature in its original tongues. Far less is known about the speakers of the Trans-Himalayan proto-language. The current volume, proceedings of a conference held in Berlin (May 2008) offers 14 studies of what Himalayan groups say of their origins and what modern scholarship currently offers on the same topic.

Although a Euhemerist approach to mythology and an ahistorical hypostasized ethnos overshadow any investigation of such themes, the papers in this book avoid these dangers. In particular, the essays of Robbins Burling (pp. 49–62) and F. K. L. Chit Hlaing (pp. 239–251) warn that “one cannot in general ask usefully where such and such a people in their current identity and under their present ethnonym were very far back in time” (Chit Hlaing, p. 239). Burling recounts how occidental researchers bring their naïvété into the field and spread it among their informants; he traces the Denkfehler of asking ‘where do they come from?’ from Sir William Jones, through the colonial officers of Northeast India, right up to a 2001 essay by Randy LaPolla (p. 58). Chit Hlaing draws attention to the emergence of the Red Karen as an ethnic identify from the late eighteenth century amid the vicissitudes of the international teak trade and describes the development of Chin and Kachin identity through reactions with low land peoples.

The essays of Geoff Childs and Toni Huber share a goal of showing that migration is a multifaceted phenomenon and one-off mass migrations of entire ethnic groups almost never occur. Nonetheless, the two essays diverge in methodology and in the interest of their findings. In Childs’ essay (pp. 11–32) the promise of showing how migrations are processes rather than events is not fulfilled. After rehearsing recent discussions in migration theory he applies these with a cookie-cutter to the history of the Sama village in Nubri valley. He offers a list of legendary discreet mass migrations, with articulated processes supplied through speculation. For example, he speculates that yogins, by maintaining wide and loose networks, helped to disseminate information about economic opportunities in distant locales, this news inspiring immigration, but he gives no evidence that the peregrinations of a specific yogin ever facilitated a specific migration.Footnote ² Discussion of “Tibet's medieval emperors” (pp. 18–19) and reporting Mi la ras pa's mythical visit to Nubri according to an unnamed and uncited hagiography as a reference to Nubri “from the late tenth or early eleventh century” (p. 18) reveal Childs to lack the Tibetological competence for the undertaking he sets himself.

To undermine the notion of migrations of whole ethnic groups en masse across vast distances Toni Huber (pp. 83–106) presents a detailed description of historically attested migrations in the northern Subansiri river catchment of Arunachal Pradesh from 1906 until today. Huber isolates seven factors that motivate migrations: local conflicts, bamboo flowerings (which can lead to explosions in the rat population, rat infestations, famine and disease), slavery, patrilocal marriage, labour migration, international conflicts, and new roads. The first four factors have occurred since time immemorial; the last three are more recent developments. The migrations Huber describes involve dozens, at most hundreds, of people and distances of a few kilometers. Huber's study relies on Tibetan, British, and Indian written sources as well as local interviews; it is a fine piece of scholarship that draws important theoretical conclusions from detailed empirical research.

Most of the remaining papers take up migration narratives as they occur in oral literature: Martin Gaenszle on the Rai (pp. 33–47), Stuart Blackburn on the Apatani, Alexander Aisher on the Nyishi (pp. 63–81), Mandy Sadan on the Jinghpaw (pp. 253–274), Charles McKhann on the Naxi (pp. 275–298). These papers range in their level of detail from the sketchy (Sadan) to the thorough (Gaenszle, Blackburn, McKhann), yet none provides the full text of the story or recitation discussed.Footnote ³ In the absence of these primary data it is impossible for future research to build on the contributions of these authors. For example, Aisher recounts a Nyishi origin myth in which an extra sun is killed by arrow (p. 66). This ‘shooting the sun’ is a key inherited myth, comparable to Indo-European dragon slaying. In Chinese it is the story of 后羿 Hòuyì (in the 山海經 Shānhǎijīng, c. 500–200 BCE, and the 淮南子 Huáinánzi, c. 79–122 BCE), in Nosu it is the tale of Zhyx ge ax lu;Footnote ⁴ the Meithei know the story as the Numit Kappa;Footnote ⁵ there is also an oral Lepcha version, unfortunately available only in English.Footnote ⁶ The insightful observations which comparison of these versions could reveal, analogous to Watkins’ disovery that Rig Vedic kap thám úd dadhātana ‘raise high the phallus’ and Lyric Greek anáget’, euruchōían (. . . poieîte) ‘stand back (make) plenty of room’ have the same scansion, are both second plural imperatives, and both occur during the presentation of a phallus (1995: 160), are not possible until the texts that contain these passages are published in full in the original.

With the achievements of Indo-European linguistics still in mind, it is disappointing that linguists contribute the least to the book under review. Nothing in the contributions of George van Driem (pp. 187–211) or Mark Post (pp. 153–186) is here published for the first time. Although unoriginal, van Driem's essay is un-problematic. In contrast, Post's paper presents problems. Post's proto-Tani-forms are nowhere explained or defended (p. 174 et passim). He postulates the erstwhile influence of a lingua franca in the history of the Tani language family without evidence (p. 164). He is likewise over-eager to see a substrate at work in Tani and Milang. Obvious cognates are at hand for many of the words he says “have scarce or non-occurring cognates elsewhere” (p. 179): Milang ŋot- ‘bite’, Lashi ŋa:t ³¹, Risianku Tamang ¹ ŋjat ‘chew’, etc.; Milang ru- ‘request’, Tib. źu < *rju (cf. Tib. źag ‘day’, Bur. ryak); proto-Tani *rjap ‘door’, Bur. ryap ‘to stand, stop’; pTani *puk ‘arrow’, Tib. √pug ‘pierce’ (pres. ḫbigs). When one considers both that Tani at times has r- where other languages have l- (e.g. pTani *pri ‘four’ versus Tib. bźi < *blʲi, OBur. liy, Chi. 四 sijH < *s.lij-s) and that as yet unaccounted for variation between -ik and -iŋ reoccurs in cognates across the family (e.g. Tib. sñiṅ ‘heart’, Lashi -), Post is mistaken to claim that proto-Tani *rɨk ‘field’ is not cognate with *liŋ (cf. Tib. źiṅ < *lʲiŋ ‘field’, Kurtöp ^Lleŋ, and Chi. 田 den < *lˤiŋ, etc.). Post discusses the Wörter und Sachen method, but refrains from engaging in it himself, with the contradictory explanations that “data published to date is [sic] insufficient” (p. 165) and “space prevents me from providing large lists of such forms here” (p. 166). He devotes ten pages (pp. 169–179) to a discussion of whether the Tani languages are more Indo-spheric or Sino-spheric, despite himself concluding in a previous paper that Matisoffs theorization of languages into a Sino-sphere and Indo-sphere is erroneous.Footnote ⁷ Why invite readers to relive this Irrweg? Despite these many oddities, it is Post's vision of the boundaries between disciplines that most shocks; he suggests that preparing “large compendia of well-transcribed, analysed and translated legends/folktales” (p. 182) is a task for others, not linguists like himself. If descriptive linguists hold themselves aloof from the documentation of traditional literature the future of comparative research on the languages of this family is bleak.