Volume 129, Issue 1

Rhotic metathesis is a characteristic feature of the northwest Middle Indo-Aryan dialect called Gandhari. At the early stage, in the Aśokan rock edicts, rC is eliminated by metathesis with either the preceding vowel or the following consonant, resulting in Cr if not simplified by r-loss. At the later stage of the Khotan Dharmapada and other Gandhari texts, metathesis in the opposite direction, from Cr to rC, is also observed. This paper evaluates contact metathesis in Gandhari by examining parallels in relevant languages and other Indo-European languages, and by scrutinizing the motivations behind the attested metathesis. The developments of r-clusters and syllabic r̥ in the Aśokan rock edicts and the Khotan Dharmapada show that in Gandhari, Cr or, at a later stage, stop-r clusters tend to be more persistent than other types of r-clusters. Different contact metatheses of r in Gandhari are a shift towards more stable sequences-or at least away from unstable sequences- and are thus linguistically plausible, although it has been asserted that some cases of metatheses are merely orthographic.

In the field of Slavonic historical phonology, it has long been known that many of the Proto-Indo-European back vowels become front after *j in Late Common Slavonic (henceforth LCS). We call this sound change Vowel Fronting. Vowel Fronting plays an important role in the rise of one of the major characteristics of the Slavonic languages, namely, the distinction between hard and soft declensions. However, not all the back vowels become front after *j in LCS, and the exact conditions for the vowel fronting have not been found yet. Here we show that the fronting is a regular sound change. The presence or absence of the etymological syllable-final consonants, which are lost in LCS, plays a decisive role in Vowel Fronting. By assuming that Vowel Fronting occurred before syllable-final consonants are lost, we define Vowel Fronting as a sound law that occurs in the following four environments without any exception: (1) in final closed syllables, (2) in light syllables, (3) between two *j’s, and (4) when the nucleus is *y [ɨ].

The Tocharian languages show an extensive array of animal names, some of which are still in want of an etymology or even a precise definition. In this paper, I undertake an in‑depth study of the Tocharian A (TA) noun śemäl, which has been translated as ‘goat’, ‘sheep’, or ‘small livestock’. First, I examine every instance of this lexeme within the Tocharian corpus to establish the precise meaning. This study confirms that śemäl refers to domestic animals regardless of their size, as well as animals used in sacrificial rituals. In the second part of the paper, I review previous etymological proposals, highlighting their strengths and weaknesses with respect to phonology, morphology, and semantics. I then propose my own account of the etymology of śemäl as an appurtenance derivative of an abstract derived from the well‑attested Proto‑Indo‑European item *ḱéi̯‑mo‑ ‘(at) home’, also reflected in Germanic (PGmc. *haimaz), Balto‑Slavic (reshaped as an ā‑stem in OCS sěmьja f. ‘family’), and, if the form to be reconstructed is actually *tḱéi̯‑mo‑, also in Indo‑Iranian (Skt. kṣema‑ m./n. ‘secure living’). TA śemäl, derived with the appurtenance suffix *‑lo‑ from an *-i-stem abstract, would have meant ‘belonging to the household’, hence the meaning ‘domestic (animal)’.

It is shown that the seemingly random alternation in the position of attributive adjectives in Old Irish can be explained by a system of referent/reference modification as first described by Bolinger (1967). The correspondence is prenominal = reference modification and postnominal = referent modification, not unlike some modern Romance languages. Comparing the Brittonic evidence, the phenomenon is found to be of Insular Celtic age, where it applied to a small class of underived prototypical adjectives.

This article is devoted to the study of potential dative subjects in Gothic, the earliest attested Germanic language, focusing in particular on word-order distribution in the Gothic Bible, the Skeireins, and the Bologna Fragment. This entails a comparison between affirmative clauses, negated clauses, and interrogative clauses, contrasting nominative subjects with potential dative subjects across both translated and native Gothic passages. We expand Ebel’s (1978) methodology to the Bologna Fragment in order to confirm which syntactic structures are native to Gothic. A comparison of the word order found with nominative subjects and potential dative subjects in native Gothic passages reveals that potential dative subjects pattern unambiguously with nominative subjects in several respects. Earlier research has documented that potential non-nominative subjects in Gothic pass the control infinitive test (Barðdal & Eythórsson 2012). Here we adduce further evidence for their subject status based on word order and a hitherto undocumented example of long-distance reflexivization.

Prohibitive μά (/maː/) stands out as one of the most remarkable features of the poorly documented Thessalian variety of Hestiaeotis (Western Thessaly). The regular outcome μεί (/meː/) (< PrGk *mē), with the characteristic raising of its long vowel, is well documented in the other Thessalian regions, making unexpected μά all the more puzzling. Attempts to explain this deviant outcome in terms of a phonetic development are largely speculative and unsatisfactory in several respects. This sort of anomaly clearly suggests that a solution should be sought outside of the realm of phonology. Hence, in this paper, the substitution of prohibitive μά for μεί in Hestiaeotis is attributed to the analogical influence of the particle μά (/ma/) used in negative oaths.

Late Khotanese phyaṣṭa‑ ‘shining, brilliant’ is from older *phiṣṭa‑, past participle of the inchoative *phīs‑ ‘to move; to shine’ from the polysemic Indo-Iranian root *tu̯ai̯š‑ ‘to be excited, be in motion, glow’ (Indo-European *tu̯ei̯s‑, Skt. tveṣ‑, Iranian *θu̯ai̯š‑). The Iranian cluster *θu̯ > *fu̯ resulted in Khotanese ph‑ initially and ‑hbetween vowels similarly to Iranian *f but with velarizing effects due to *u̯ before its disappearance. *θu̯‑ > ph‑ is observed in eight words with so far uncertain etymology, all derived from *phīs‑. Scrutiny of the occurrences of the derivatives and reinterpretation of some of them confirm the continuation of the two senses of Indo- Iranian *tu̯ai̯š‑ in Khotanese *phīs‑. Khotanese is thus the only Iranian language to continue Indo-Iranian *tu̯ai̯š‑ also in verbal formations: phyaṣṭa‑ ‘shining, brilliant’; śśī-phīsa‑ ‘whitish’ ← ‘approaching white’ or ‘having the brightness of white’; phīsūna‑ ‘(symbolic hand) movement, gesture’; usphīs‑ ‘to emerge; to shine forth’ (with us‑ ‘up, out’); *usphīsa‑ ‘upward movement’ in usphīsaroña‑ ‘arrogance’; iterative phīś‑ ‘to escape from, avoid’; iterative naṣphīś‑ ‘to depart from’; and causative haṃphīś‑ ‘to combine’ ← ‘to cause to go together’ (in addition to pūheiʾta‑ ‘slim, slender’ < *pati-θu̯ai̯š-a‑).

Several PIE forms with a word-final sequence *-ō̆m would be morphologically better understandable if they ended in *-ē̆m. It is therefore proposed that, in its prehistory, Proto-Indo-European underwent a sound law *-ē̆m > *-ō̆m. This article will treat the relevant evidence in favor of this new sound law, as well as discuss an apparent counterexample. Moreover, it will offer some typological parallels for this development.

This paper examines the use of the characters t h eta ⟨θ⟩ and k h i ⟨χ⟩ in Transalpine Celtic inscriptions engraved in Greek characters. Since /t h / and /k h /, respectively, were not needed for Celtic, it is conventionally presumed that these characters were repurposed in these inscriptions, t h eta ⟨θ⟩ as the tau Gallicum phoneme and k h i ⟨χ⟩ as the allophone [x] of the phoneme /k/ before /t/. It is argued that, in fact, the aspirated voiceless plosives of ancient Greek had become fricatives in southern Transalpine Gaul before the Celts of the region started employing Greek characters to engrave inscriptions in their language, and thus that t h eta ⟨θ⟩ and k h i ⟨χ⟩ in Transalpine Celtic inscriptions were employed with their proximate phonemic values in Greek. We thus can determine a proximate terminus ante quem for the frication of the ancient Greek aspirated voiceless plosives in southern Transalpine Gaul, which leads to the prospect for Transalpine Celtic having contrasted its plosive series not via the Laryngeal feature [voice], but via [spread glottis].

This article deals with a number of Middle and New Iranian words of doubtful etymology for which a derivation from Ir. *hakat ‘together, at once’ is proposed. Its possible reflexes in Western Iranian witness a development towards a conditional or temporal meaning (Parth. ag ‘if’, NP aknūn ‘now’), but remnants of the old meaning ‘together’ can be identified in MP ag(e)nēn ‘together’ and in the Arm. loanword hakaṙak ‘opposite, against’. In Eastern Iranian it seems to have undergone grammaticalization and developed into a preverb (Khot. gga‑) or a preposition (Sogd. ku ‘towards’, Orm. ku- ‘to, on, etc.’, Ishk. kI ‘to, on, for’, Zeb. ka ‘id.’). Especially in this latter group of languages, Ir. *hakat shows an evolution largely comparable to that of *hačā ‘from’, which, according to most scholars, belongs to the same root.

In this article, I argue that the counterfactual function of the optative was inherited from Proto-Indo-European and that its past-referring counterfactual function (e.g. in Vedic Sanskrit, Archaic Greek, and Gothic) seems to be the oldest preserved function of the optative in early Indo-European languages. In these and in languages where the optative underwent mood syncretism with other moods such as the subjunctive (e.g. Latin or Old English), the inherited counterfactual function extended its temporal reference to the non-past, following the life cycle of counterfactual markers. Over time, this mood, syncretized or not, lost its counterfactual function in most Indo-European descendants and was replaced by different mood formations at various speeds (typically by a past indicative or a past-like ‘conditional’ mood). First, I provide a functional history of the optative in Ancient Greek, from (1) past counterfactual to (2a) non-past counterfactual, and from there into (3a) non-past ‘hypothetical’ usage and from (1) past counterfactual into (2b) past habitual and (3b) past generic. Next, I internally reconstruct the developmental paths of the optative and its counterfactual replacements across Indo-European languages, concluding with a tentative diachronic typology of counterfactual mood strategies.

This paper presents evidence for a regular sound change in 12th-century Old French where word-initial /k/ voiced to /ɡ/ before /la/ or /ra/. My treatment is the first to propose a unified account, as due to a regular sound change, for 13 reflexes of inherited etyma and 4 reflexes of Germanic borrowings that all exhibit velar onset voicing (VOV) contemporaneously. A dating in the late 12th century for VOV is supported by both corpus evidence and how it is bled, counterbled, and fed by other sound changes whose dating is established in prior literature. The ultimate origins of this Old French velar onset voicing most likely lie in an older and phonetically conditioned synchronic voicing process that originally operated both word-internally and at boundaries. Word-internally, where conditioning was static, the alternation became a phonemic difference and thereafter lost productivity relatively early. But synchronic alternation continued at boundaries, where its conditioning continued to depend on the adjacent word. It survived in a progressively reshaped manner word-initially, until the late 12th century, when this voicing alternation, now much narrower in scope, gave rise to VOV as a regular diachronic sound change. Previous treatment of the words in question considered each only in isolation, bypassing the chance to consider the proposed regular sound change as a possibility. Thus, this study shows the importance of considering both lexically focused approaches and the probing for regular sound changes as necessarily complementary methods.

A characteristic trait of Vedic as well as Classical Sanskrit is the use of nominal compounds. Diachronic linguistic studies have observed an increasing use of compounds in Vedic texts. It is also generally accepted that compounds should be read as syntactic phrases and that they can be equivalent to subordinate clauses. However, it has not been studied so far whether and to which degree compounds replaced competing syntactic structures such as relative clauses or participial constructions over time. Using data from a syntactic treebank of Vedic and early Classical Sanskrit, this paper addresses the questions whether compounding replaced equivalent constructions and which textual and sociolinguistic factors may have driven this process. The paper studies compounds used as adnominal and adverbial modifiers, and compares their frequency distributions with those of subordinate clauses, adjectives, converbs, and participial constructions. Since the number of relevant cases is limited and the sociolinguistic factors driving the use of compounds are not well understood, the observed distributions are modeled with a hierarchical Bayesian framework that extracts an optimal subset from a set of possible explanatory factors (chronology, geography, poetry/prose alternation, genre, and school affiliation of Vedic texts).