Ampābru

To serve as my first serious a posteriori conlang, I introduce Ampābru, a romance language born out of the dying African Latin whose speakers in my alternate timeline were transferred to western India by the year 700 CE, whence came great influence thereupon.

Phonology

The maximum permitted syllable structure is C₂VC₃ in most cases, although some may extend to C₂VC₄. Onset clusters cannot begin with /s/, /ɳ/, /ŋ/, nor /ɽ/. /𝼈/ is forbidden from appearing anywhere within the onset.

Herebelow lies the phonemic inventory of consonants:

		Labial (L)	Alveolar (A)	Retroflex (R)	Velar (V)	Uvular (U)
Nasal		m	n	ɳ	(ŋ)
Plosive	voiceless	p	t̪	ʈ	k	q
	aspirated	pʰ~ɸ	t̪ʰ	ʈʰ	kʰ	(qʰ)
	voiced	b	d̪	ɖ	g
	murmured	bʱ~β	d̪ʱ	(ɖʱ)	gʱ
Fricative			s	ʃ~ʂ		ɦ
Liquid	median		ɾ~ɹ	ɽ~ɻ	j
	lateral		l̪~ɬ	𝼈

It ought be noted that /j/ is in fact palatal and not velar.

The following comprises the vowel inventory:

	Front		Back
Close	i	y	(ɨ)	u
Close-Mid	e		o
Open-Mid	ɛ	ə	ɞ	ɔ
Open	æ	a		(ɒ)

Orthographically (for Latin script), if ń or ḿ appear, this indicates that the preceding vowel is nasalised, and intervocallically, the /n/ or /m/ phoneme is reinserted via sandhi — this does NOT occur for /ɳ/ since it resisted nasalising surrounding vowels. Should a tilde diacritic be used instead, there shall be no such sandhi and the vowel is merely to be nasalised. The Devanagari equivalent to this convention is to use an anusvāra whereinever sandhi is applicable herein but use a chandrabinďu whereinever it isn't but the vowel is nonetheless nasalised.

Certain younger speakers of Ampābru, though particularly not those of the diaspora, have begun to merge /ɒ/ with /ɔ/ and /y/ with /i/, as notated below. The former's nonpresence for many diasporic speakers may have resulted from resistance to the North American COT-CAUGHT and/or FATHER-BOTHER vowel merger.

ɒ→ɔ/_

y→i/_,!R_

y→ɞ(ɪ)/R_

/ɨ/ isn't per se its own phoneme, and is not transcribed in orthography. It is only inserted to variable degree (never intervocallically) to smooth out clusters starting with aspirated or murmured consonants, as in qhčābra (narrow: [qʰʁ̥a.brə], dn: ख़्चाब्र), meaning "to escape", from Latin escapare.

As notated below, /pʰ/ and /bʱ/ change realisation intervocallically, with similar occasionally happening to /t̪/ and /d̪/:

[+stop,+asp/mur,+labial]→[+fric]/V_V|V∉{close},σ_CV

[+stop,+asp/mur,+dental]→[+fric]/V₁_V₂|V∉{close∪close-mid}∩V₁=V₂

The consonant transcribed in Latin script as č, čh, j, jh are notoriously variable between dialects. They are in Devanagari represented by the "palatal" series च, छ, ज, झ, but in reality may have their primary realisation as [k(ʲ)~ʈ(ʲ)~ʈɽ̊(ʲ)~ɻ̊˔~ʁ̥], [k(ʲ)ʰ~ʈ(ʲ)ʰ~ʈɽ̊(ʲ)ʰ~ɻ̊ʰ˔~ʁ̥ʰ], [g(ʲ)~ɖ(ʲ)~ɖɽ(ʲ)~ɻ˔~ʁ], and [g(ʲ)ʱ~ɖ(ʲ)ʱ~ɖɽ(ʲ)ʱ~ɻʱ˔~ʁʱ] respectively, though none of these are most common. However, the vast majority of their appearance is from loanwords from surrounding Indo-Aryan languages who possess post-alveolar affricates, which aren't present herein and were reanalysed by the speakers to fit better with Ampābru's phonology.

Evolutionwise, earlier vowel shifts are modelled off of Sardinian, whereas later ones take influence from Hindi, Marwari, and Gujarati.

The following table lists the changes in vowels during Ampābru's development:

Classical 200s	Afr. V. Latin year	Far East Afr. Romance year		Old Ampābru 1100s		Middle Ampābru 1500s		Mod. Ampābru 1900s+
		close	open	close	open	close	open	close	open
a(ː)	a	a		ɔ		ɔ		ɔ
a͜e	ɛ͢e	ɛ	e	ɛ,(ə)	e	ɛ/ə	e	ɛ/ə	e
o͜e
e(ː)	ɛ	ɛ		ə	a͜ɪ͜/aʊ,e	ə	a͜ɪ͜~æ/aʊ~ɒ,e	ə	æ/ɒ,e
i/y(ː)	i	i		i		i		i
o(ː)	ɔ	ɔ		ɔ		ɔ		ɔ
e͜u	ɛ͜w	ø	e͜ø	ɪ͜ɜ	i͜y	ɜ	ʏ͜y	ɞ	y
a͜u	ɛ͜o	o		o		o		o
u(ː)	u	u		u		u		u

---

Orthography

Latin-script orthographywise, Ampābru's is of a comparable up-to-dateness to a language like Spanish, retaining written qu, gu, and c. It is important to note that k is more common in Indic loanwords, whereas c, which has retained its pronunciation as k in the vast majority of places similarly to what is evidentially known of African Romance and also even Sardinian in the present day, whereto Ampābru can be said to be closeliest related.

Here is a table of the Latin orthography:

Here is a table of the Devanagari orthography, wherein the c-k, qu-kh, and similar distinctions are dropped:

---

Grammar

From its extreme isolation from other Romance languages and its great contact with morphologically dense languages such as Tamazight and Hindi (albeit less so than Tamazight), there has been substantial borrowing from these substrates both lexically and in some ways even grammatically.

For instance, word order is primarily SVO, although there exists some flexibility due to noun case marking. Clusivity distinction in the first person between 1+3 and 1+2(+3) also developed from Gujarati influence. Ergativity is largely ergative-absolutive, a dramatic departure from Latin's nominative-accusativeness.

It should be noted that the genitive case in pronouns is falling out of use in favour of the dative.

The following are the personal pronouns.

	Personal Pronouns									Interrogative/Relative
	1st Person			2nd Person			3rd Person
	Singular	Plural		Singular		Plural	Singular		Plural	Singular		Plural
		1+3	1+2(+3)	intimate	formal		masc.	fem.		anmt.	inan.
Ergative	akh (kh')	naš	āp	tu (t')	taḿ	vaš	aš (š')			qùi	qùaìn	qùōhiyē
Absolutive	mē		āpē	tē			lu	lā	li
Vocative	maìn			taìn			laìn
Dative	mēi	nadh		tēi		vadh	lii/lēi	lēi		bhi		bhiš
Genitive	mi/mēi/mēr			ti/tēi/tēr	tmhār		li/lãìr	lē/lãìr
Inessive	cũmku	cãnšku	-	cutku	-	caùšku	cũšku
Reflexive	mē (m')	naš		tē (t')		vaš	sē (s')

Verb conjugation is extensive, with the notable development of a presumptive mood.

Tense		Non-Past													Past
Aspect		Perfective							Continuous						Perfective							Continous						Habitual
Number		Singular			Plural				Singular			Plural			Singular			Plural				Singular			Plural			Singular			Plural
Clusivity					1+3			1+2(+3)										1+3			1+2(+3)
Mood	Stem	‑(r)a ‑eùr	‑(r)ā	‑(r)i	‑(r)a ‑eùr	‑(r)ā	‑(r)i	any	any			any			‑(r)a ‑eùr	‑(r)ā	‑(r)i	‑(r)a ‑eùr	‑(r)ā	‑(r)i	any	‑(r)a ‑eùr	‑(r)ā	‑(r)i	‑(r)a ‑eùr	‑(r)ā	‑(r)i	‑(r)a ‑eùr	‑(r)ā	‑(r)i	‑(r)a ‑eùr	‑(r)ā	‑(r)i
Indicative	1st Masc.	‑ā		‑ya	‑amha	‑āmha	‑imha		‑rē			‑reyā			‑aì			‑āmha		‑imha		‑rētā			‑retē			‑tā			‑tē
	1st Fem.	‑i		‑hi											‑e							‑rēti						‑ti
	2nd Intimate	‑aš	‑āš	‑yaš	‑atha	‑ātha	‑itha								‑āštta		‑ištta	‑āšttša		‑išttša		‑rēta						-ta
	2nd Formal	‑aù		‑yō	‑ẽ		‑hẽ		‑raù									‑raùnta		‑reùnta
	3rd Masc.	‑ā		‑ya	‑anta	‑ānta	‑eùnta		‑rē						‑a	‑ha						‑rētā						‑tā
	3rd Fem.	‑i		‑hi																		‑rēti						‑ti
Imperative	1st Masc.	‑anna	‑ānna	‑yanna	‑ām			‑āḿp							‑hēnna		‑yēnna	‑aìm			‑aìpnne
	1st Fem.	‑anni		‑yãi											‑hē̃i		‑yē̃i
	2nd Intimate	‑a	‑ā	‑ya	‑ath	‑āth	‑ith								‑hē		‑yē	‑hēth		‑yēth
	2nd Formal	‑aù			‑anha	‑ānha	‑eùnha								♂:‑a ♀:‑aì	‑šrēnha
Subjunctive	1st Masc.	‑ā	‑ya		‑āmha	‑yamha			‑rē			‑remhē			‑ri	‑si		‑rimha	‑simha			‑rēt			‑rēmhat			‑ra			‑ramhē
	1st Fem.	‑aì	‑yaì
	2nd Masc.	‑āš	‑yaš		‑ātha	‑yatha						‑rethē			‑šri	‑šši		‑ritha	‑sitha						‑rēthat			‑šra			‑rathē
	2nd Fem.		‑yaì
	3rd Masc.	‑ā	‑ya		‑ānta	‑eùnta			‑rō			‑reùntē			‑ri	‑si		‑rinta	‑sinta			‑rōt			‑rant			‑ra			‑rēń
	3rd Fem.	‑aì	‑yaì
Presumptive	1st Masc.	‑ugā		‑uìgā	‑ugē		‑uìgē		‑rugrepha			‑ruremhē			‑ugphē		‑eùgphē	‑umhē		‑uìmhē
	1st Fem.	‑uìgā			‑uìgē
	2nd All	‑ugā		‑uìgā	‑thugā		‑thuìgā					‑rudhrephē						‑udhphē		‑uìdhphē
	3rd Masc.	‑ogā			‑ongā				‑rogrepha			‑roreùmhē			‑ogphē			‑omhē
	3rd Fem.	‑eùgā
Counterfactual	1st Masc.	‑at		‑yat	‑aìmha	‑amha	‑uìmha		‑trē			‑treyē			‑t			‑te				‑trat			‑treyat			‑trā			‑trē
	1st Fem.	‑aìt		‑yaìt											‑ta													‑tri
	2nd All	‑at		‑yat	‑aìtha	‑atha	‑uìtha								‑t													‑tra
	3rd Masc.				‑aìnta	‑anta	‑ũìta																					‑trā
	3rd Fem.	‑aìt		‑yaìt											‑ta													‑tri