On the position of Dameli

Dameli is an Indo-Iranian language whose precise classification as Indo-Aryan or Nuristani is still debated. Indeed, Dameli contains a substantial layer of Nuristani-derived words, some of which are part of the language's basic vocabulary. Here, we will try to sort out the Indo-Aryan and Nuristani layers on both lexical and morphological grounds.

(For abbreviations and sources, see this post. For Nuristani orthographic conventions, see this post.)

Documentation
Morgenstierne (1942) provides us with the first detailed look of the Dameli language, calling it a "Kafir-Dardic" language. He describes some salient Nuristani phonological developments (such as the development of IE palatovelars) alongside a sketch morphology and lexicon.

Perder (2013) is the first comprehensive grammar of Dameli, covering most essential features of the language. We will use Perder's orthography when citing Dameli forms.

The Nuristani component of Dameli
One easily recognizable criterion for identifying a Nuristani-derived word in Dameli is the development of Proto-Indo-Iranian palatals. While Indo-Aryan has *ć > ś and shows a merger of *ȷ́(ʰ)/*ǰ(ʰ) as *j(ʰ) > j/h, the Nuristani stratum shows a development *ć > ċ /ts/ and *ȷ́(ʰ) > z, *ǰ(ʰ) > ž. Much of these salient details were already summarized in Halfmann (2022, 2025).

Examples:
PIIr. *ȷ́ʰr̥d- ~ *ȷ́ʰr̥dai̯a- "heart" > Dam. zaadi
PIIr. *ǰʰan- "to kill" > Dam. žan-
PIIr. *ā-ga-sća- "to come" > Dam. aaċ-
PIIr. *ȷ́āmātar- "son-in-law" > Dam. zaamaa
PIIr. *ȷ́ā-naH- "to know" > Dam. zaan-
PIIr. *ću̯an- "dog" > Dam. ċuna
PIIr. *kać- "to look, show" > Dam. kaċ-
PIIr. *tać- "to carve" > Dam. taċ-
PIIr. *ȷ́ʰai̯ana- "winter" > Dam. zin

The word daš(t) "hand" < PIIr. *ȷ́ʰasta- must also belong to the Nuristani-derived lexicon, since it reflects an irregular sound change to *d (which is also shared with Iranian except Avestan).

Another feature found in the Nuristani component is the development of *st/*št, which is always št (> š word-finally) and RUKI *š developing like Nuristani *s, *š (/i_, /_i).

Examples:
PNur. *prošta- "bed" > Dam. proš(t)
PIIr. *ȷ́ʰasta- > *dásta- > Dam. daš(t)
PIIr. *tuša- "straw" > *túsa- > Dam. thus (secondary aspiration)
PIIr. *stuHnaH- "pillar" > Dam. uštuun
PIIr. *ni-šid- > Dam. niš-

Dam. kaar "ear" < PIIr. *karna- shows the Nur. development, with lengthening of *a before *rn > r.

The Indo-Aryan component of Dameli
These show the OIA developments mentioned above, as well as RUKI *št > OIA ṣṭ and some developments peculiar to later stages of IA, such as geminate *kk in "one" and sv > *sp,  retention of *s before *i/*y, and retention of short a before rC.

Examples:
PIIr. *Hai̯ka- "one" > OIA éka- > MIA *ekka- > Dam. ek
PIIr. *dʰugʰHtar- "daughter" > OIA dúhitṛ- > MIA *dhitu > *dhiu > Dam. žu
PIIr. *darć- "to see" > OIA dṛś- > Dam. dac̣- "to look"
PIIr. *su̯aćura- "father-in-law" > OIA śváśura- > MIA *śpaśura- > Dam. pašur
PIIr. *daća "ten" > OIA dáśa- > Dam. daš
PIIr. *su̯asar- "sister" > OIA svásṛ- > MIA *spásā > Dam. pas
PIIr. *aȷ́ʰam > OIA ahám > Dam. ay
PIIr. *su̯asri(H)i̯a- "nephew" > OIA svasrī́ya- > MIA *spasriya- > Dam. ṣpaṣi
PIIr. *tasi̯a "of that" > OIA tásya > Dam. tas
PIIr. *Hu̯iHćati "twenty" > OIA *vīśatí (attested viṃśatí) > Dam. biši
PIIr. *marti̯a- "mortal" > OIA mártya- > Dam. mač
PIIr. *manušii̯a- "man" > OIA manuṣyà- > Dam. maaṣ "people"
PIIr. *Hr̥ćša- "bear" > OIA ṛ́kṣa- > Dam. iṇc̣
PIIr. *diȷ́ʰu̯aH- "tongue" > OIA jihvā́- > MIA *jibbhā- > Dam. žip
PIIr. *HaHs- "mouth" > OIA āsyà- > Dam. aas
PIIr. *ćr̥-nau̯- > OIA śṛṇóti > Dam. uštruun- (epenthetic u and t)
PIIr. *pantaH- "path" > OIA pánthā- > Dam. phan(t) (aspiration metathesis)

Dameli pronouns
In an attempt to identify the core layer of Dameli as Indo-Aryan or Nuristani, a critical examination of pronominal forms will be made, as they form part of the lexicon most resistant to borrowing.

1sg.nom. ay
Notes: This shows a clearly Indo-Aryan development of *ȷ́ʰ > h with subsequent loss. While the other Nur. forms indicate PNur. *j̈ (eg. Kt. (W) vúze ~ vúzë, Pr. unzˈū), forms such as Ash. ay may point to an Indo-Aryan borrowing. The glide, however, remains unexplained.

1sg.obl. muu
Notes: Clearly derived from OIA ma-. The Nur. languages appear to reflect loss of *m (and occasional prefixation of *i-) with a possible exception in Prasun. The preservation of m is typical across the IA and Ir. languages spoken within the region, so this word must belong to the Indo-Aryan layer.

1pl.nom. ay
Notes: This form is identical to the 1sg.nom. and may have replaced a reflex of OIA vayám found in other IA languages but not in Nur.

1pl.obl. amaa
Notes: Corresponds to OIA gen. asmā́kam. Nur. cognates appear to show prothetic *i- (Ash. (S) imë, Kt. (NE) imú, NKal. (W, Z) yama) except for Prasun. Since it is not found in this word, it is more likely to belong to the Indo-Aryan layer.

2sg.nom. tu
Notes: Corresponds to OIA tuvám. The phonetic developments do not tell us whether the word is Nuristani or Indo-Aryan.

2sg.obl. too
Notes: Corresponds to OIA gen. táva. The phonetic developments do not tell us whether the word is Nuristani or Indo-Aryan.

2pl.nom. bi
Notes: Corresponds to Ash. vi, NKal. vi, Tr. vi, Shum. wī, and possibly Kho. bisa. Since it is only found in the "Southern" Nuristani languages, which are more influenced by Indo-Aryan, it is possible that the form belongs to the Indo-Aryan layer, though the assumption of a Nuristani layer would be equally plausible.

2pl.obl. mya
Notes: Corresponds to OIA yuṣma-. The glide may come from a metathesis of *yma, and does not necessarily indicate a Nuristani prothetic *i-. Within Nur., only Prasun shows m (müo ~ uman) and since similar forms appear in the neighboring IA languages, this word likely belongs to the Indo-Aryan layer.

3sg.anim.nom. i (prox.)/see (dist.)
Notes: Corresponds to OIA ayám and sá and closely resembles NKal. yi and sa. The phonetic developments do not tell us whether the word is Nuristani or Indo-Aryan.

3sg.anim.obl. mas (prox.)/tas (dist.)
Notes: Corresponds to OIA amúṣya and tásya. The form mas is apparently secondary, implying an intermediate *amasya remodeled after tásya. The lack of palatalization before a glide indicates that this word belongs to the Indo-Aryan layer (against NKal. (Z) tašo).

3sg.anim.erg. manii (prox.)/tanii (dist.)
Notes: Shows a suffixal -nii. This is probably cognate to Hindi -ne and if true, would put it within the Indo-Aryan layer.

3sg.inanim. yee (prox.)/see (dist.)
Notes: Related to the animate pronoun set via an additional -ee (probably extracted from instr. téna).

Verbal conjugation
Kreidl (2025) explores the origin of the Dameli verbal endings. One salient characteristic that stands out is the derivation of Dam. 2sg.fut. -es from OIA -asi, which shows no palatalization against Nur. -š (Ash. -s shows regular *š > s), hinting at the IA derivation of these suffixes.

Aspiration features
Unlike the Nuristani languages but found in nearby IA languages, Dameli possesses aspirated consonants. However, they only occur with voiceless stops. It is easy to show from Dam. uštuun "pillar" ~ OIA sthū́ṇā- that the Nuristani layer appears to have lost aspiration, even in the voiceless context.

As mentioned by Halfmann (2025), the aspiration in the Nuristani-derived thus "straw" is secondary, and is irrelevant to the conclusion above.

The Indo-Aryan layer appears to show the loss of voiced aspiration, a natural tendency in the Hindu Kush IA languages. An example is bhāṣā- "language" > Dam. baaṣa.

Voiceless aspirates, however, appear to be preserved in the Indo-Aryan layer.

Examples:
OIA múkha- "face" > Dam. mukh
PIIr. *staH- "to stand" > OIA sthā- > Dam. th- "to be"
PIIr. *śći-nad- "to cut" > OIA chinátti > Dam. čhin-
OIA phālayati  "to split" > Dam. phaal-

On the position of the Nuristani component in Dameli
There is a clear indication that the Nuristani-derived words cannot be associated with any of the attested Nuristani languages, as mentioned in Halfmann (2022). A careful examination of the historical phonology reveals several conservatisms that are not found in modern Nuristani.

The details below are from Halfmann (2022), with some modifications:
- PNur. *a and *ā are retained in all positions
- PNur. *u is not fronted at all
- PNur. *r- is retained (though this may be a loan adaptation, as all Nur. languages show lenited reflexes, necessitating PNur. *ř-. Some apparently exceptional Pr. forms such as ürǰˈug "light" may be secondary reversals)
- PNur. *-n- is retained (shared with Pr.)
- PNur. *-r̥d- > -d- (shared with Ash.)

The last detail (PNur. *r̥d- > -d-) may point to a closer affiliation with Ashkun, though it does not share the diagnostic change of *ċ, *š > Ash. s and PNur. *č/*ǰ > Ash. ċ/j̈. In any case, the Nur. component represents a phonologically conservative offshoot of the core Nuristani continuum, indicating that the component was already present in Dameli since the latter part of the 1st millennium, before the "Central Nuristani Vowel Shift" occurred. One may compare this to Eng. king ~ Fi. kuningas, where the latter borrowed form is almost identical to PGmc. *kuningaz.

Conclusion
While there exists a substantial portion of Nuristani-derived vocabulary in Dameli, most of it is limited to the lexicon, despite its presence in several basic vocabulary items. A careful analysis of the core morphology clearly shows several points of agreement with the Indo-Aryan layer. The consistent preservation of voiceless aspirates in the Indo-Aryan layer versus the lack of it in the Nuristani layer (if we exclude secondary aspiration from sibilants) also adds support to the Indo-Aryan affiliation of the language. If Dameli was a Nuristani language under heavy Indo-Aryan influence (as in Tregami), aspiration would not have been present in the language.

Thus, Dameli can be explained as an Indo-Aryan language heavily influenced by an early, conservative offshoot of Nuristani sometime within the last 1,500 years. It is hoped that future research on this language may provide additional support to the Indo-Aryan classification of Dameli.

References

• Morgenstierne, Georg. (1942). Notes on Dameli, a Kafir-Dardic Language of Chitral. Oslo, Norway: Norsk Tidsskrift for Sprogvidenskap 12.
• Perder, Emil. (2013). A Grammatical Description of Dameli. PhD dissertation. Stockholm, Sweden: Stockholm University.
• Halfmann, Jakob (2022). Advances in the historical phonology of the Nuristani languages. International Journal of Diachronic Linguistics and Linguistic Reconstruction, vol. 19.
• Halfmann, Jakob (2025). The Diversification of Indo-Iranian and the Position of the Nuristani Languages. Wiesbaden: Reichert.
• Kreidl, Julian (2025). A Historical-Comparative Grammar of Nuristani. Leiden: Leiden Summer School of Languages and Linguistics.