User:Selguha/Lido phonotactics

This page contains drafts for a loglang/auxlang project I am currently calling Lido (for 'Lojban-ido'). Lido has barely any syntax or lexicon yet; it is being built from the phonology up.

Background

Lido has the following inventory. Throughout the rest of this document, Lido orthography rather than IPA will be used.

Lido consonant graphemes and phonemes

		Labial	Alveolar	Palatal	Velar	Glottal
Plosive	unvoiced	p	t	c  t͡ʃ	k
	voiced	b	d	j d͡ʒ~ʒ	g
Fricative	unvoiced	f	s	x ʃ	h x~h
	voiced	v	z
Nasal		m	n
Lateral			l
Rhotic			r
Semivowel		w		y j

The phoneme j (d͡ʒ) is exceptional, in that it has two primary allophones in the standard dialect. It is a fricative in coda, after heterorganic consonants and, perhaps, intervocalically. Elsewhere it is an affricate; namely, word-initially (when not extrasyllabic) and after n, l and r. Why this behavior? Purely an Anglophone aesthetic preference, and for parallelism with c/x (one may imagine that a merger occurred at some point in the past for /d͡ʒ/ and /ʒ/, but only partially affected /t͡ʃ/ and /ʃ/). It is equally standard to use only one allophone of the two.

Semivowels are not shown on the charts below. This is because clusters with semivowels are written, and may be analyzed, as involving rising diphthongs; that is, with the semivowel as part of the syllable nucleus. However, qw and qy may be declared legal heterosyllabic clusters in Level III. To be determined.

Lido employs a tiered phonological system, with multiple levels representing points along a scale from 'foreign and unassimilated' to 'native.' There are three primary levels: I, II and III. These pertain to the following lexical classes: names, culturally specific loanwords, native words. Each level has its own phonotactics, along with other constraints.

Such a baroque system (which this undeniably is) is sure to receive criticism. I believe that it is necessary in order to balance faithfulness to source languages with the desired restrictiveness, that is, simplicity, of this auxlang-like phonology. However, this argument has yet to be fleshed out, and I am quite open to being persuaded that certain aspects of the tiered system are unnecessarily complex.

While Level III phonology is much more restrictive than English, all levels are based heavily on English. This is partly because of laziness, partly by design. English is the native language of the author, hence the easiest model to for me to use. But English is also the most widely spoken language in the world after Standard Chinese. Its phonotactics are well-known to a large number of people around the world, and by far more well-known than those of any language with a comparable syllable structure. Russian, Persian, Hindi and Arabic phonotactics are the only other models worth considering. Persian and Arabic can be ruled out right away as outliers in the clusters they permit word-medially and -finally. Russian and Hindustani are both good models, but these languages were deemed too 'un-international' to justify the level of research that would be required of the author in order to base Lido phonotactics on them.

All this said, Russian and Persian phonotactics were given a cursory study during the development of Level I.

Lido's Level I phonotactics are essentially those of English, but slightly less restrictive in some ways and more restrictive in others. Aside from this, the principal way in which Lido phonotactics is English-like its preference for coronal consonants in clusters; and for heterosyllabic clusters to be of the pattern peripheral-coronal, with the coronal consonant coming second.

Level I: Names

This is numbered as the first level for two reasons: because it represents a 'baseline' of what is phonologically 'intelligible' or parsable to the prototypical Lido speaker, and because it has the fewest constraints (but the greatest variety of patterns).

ALTERNATIVELY, levels are numbered in the opposite order, with native words as Level I or Level Zero. This might prove to be more intuitive, especially since closed-class native words are conceptually a sort of endpoint, whereas foreignness seems to have no endpoint.

Level I phonotactics are based heavily on English. Syllable structure is maximally CCCVVCC, as exemplified by streinj 'Strange'. Legal coda clusters come from a subset of Level III heterosyllabic clusters: those of falling sonority or with a sibilant as the second consonant (excluding j). Names with three or more consonants in coda can be assimilated in several ways. Onset clusters are generally preferred to coda clusters, so a cluster like nstr should be syllabified as n.str rather than ns.tr. If an epenthetic vowel is needed it may be an echo vowel, e or i. More to be determined.

The first consonant of a word is able to syllabify with a vowel from the preceding word. So, lai streinjlav 'what-is-named Strangelove' is pronounced [laɪ̯‿s.ˈtreɪ̯nd͡ʒ.lav]. (This cannot cause misparsing, because CVVC is not a legal syllable in Level I words.)

Stress may occur in four patterns: penultimate, ultimate, antepenultimate or preantepenultimate. Ultimate and penultimate are tied as the least marked patterns.

Level II: Loanwords

This level corresponds to Lojban's zi'evla word-class. It pertains to words whose meaning is too specialized or culturally specific for them to have native root-word forms.

All words have final vowels; stress is penultimate. There are a few exceptional ways in which Level II is more constrained than particular sublevels of Level III.

Level III: Native words

This level includes three sub-levels for the three primary classes of Lido word: compound words, root words and function words. These are numbered IIIa, IIIb and IIIc respectively, indicating an increasing number of phonological constraints as maximal nativeness is approached. In general, Level III is characterized by relatively small sets of possible onsets and heterosyllabic clusters, and a maximal syllable template of CCV(V|C).

All words have final vowels and penultimate stress, except for 'emphatic' allomorphs of function words. These latter word-forms have a final glottal stop and ultimate stress. (Elsewhere, glottal stop is not phonemic, and its status as a phoneme in the language is marginal.)

Heterosyllabic clusters

Figure 1: Word-medial clusters of Level III

	p	b	f	v	m	t	d	s	z	n	l	r	c	j	x	k	g	h
p						pt		ps		pn	pl	pr			px
b							bd		bz	bn	bl	br		bj
f						ft		fs		fn	fl	fr			fx
v							vd		vz	vn	vl	vr		vj
m	mp	mb	mf	mv		mt	md	ms	mz	mn	ml			mj	mx
t								ts				tr
d									dz			dr
s	sp				sm	st				sn	sl					sk
z					zm		zd			zn	zl
n						nt	nd	ns	nz		nl		nc	nj		nk	ng	nh
l	lp	lb	lf	lv	lm	lt	ld	ls	lz	ln			lc	lj	lx	lk	lg	lh
r	rp	rb	rf	rv	rm	rt	rd	rs	rz	rn	rl		rc	rj	rx	rk	rg	rh
c
j					jm		jd			jn	jl	jr
x	xp				xm	xt				xn	xl	xr	xc			xk
k						kt		ks		kn	kl	kr			kx
g							gd		gz	gn	gl	gr		gj
h						ht		hs		hn	hl	hr			hx

Key

Color	Meaning
Light green	Level III heterosyllabic clusters that are also Level I coda clusters
Light blue	Other Level III heterosyllabic clusters
Indigo	Level III onset clusters (may occur intervocalically)

Figure 2: Word-medial clusters of Level II [WORK IN PROGRESS]

	p	b	f	v	m	t	d	s	z	n	l	r	c	j	x	k	g	h
p			pf		pm	pt		ps		pn	pl	pr	pc		px	pk		ph
b				bv	bm		bd		bz	bn	bl	br		bj			bg
f	fp				fm	ft		fs		fn	fl	fr	fc		fx	fk		fh
v		vb			vm		vd		vz	vn	vl	vr		vj			vg
m	mp	mb	mf	mv		mt	md	ms	mz	mn	ml	(mr)	mc	mj	mx	mk	mg	mh
t	tp		tf		tm			ts		tn	tl	tr				tk		th
d		db		dv	dm				dz	dn	dl	dr					dg
s	sp		sf		sm	st				sn	sl	sr				sk		(sh)
z		zb		zv	zm		zd			zn	zl	zr					zg
n	np	nb	nf	nv	nm	nt	nd	ns	nz		nl	(nr)	nc	nj	(nx)	nk	ng	nh
l	lp	lb	lf	lv	lm	lt	ld	ls	lz	ln		(lr)	lc	lj	lx	lk	lg	lh
r	rp	rb	rf	rv	rm	rt	rd	rs	rz	rn	rl		rc	rj	rx	rk	rg	rh
c	cp		cf		cm	ct				cn	cl	(cr)				ck		ch
j		jb		jv	jm		jd			jn	jl	jr					jg
x	xp		xf		xm	xt				xn	xl	xr	xc			xk		xh
k	kp		kf		km	kt		ks		kn	kl	kr	kc		kx			kh
g		gb		gv	gm		gd		gz	gn	gl	gr		gj
h	hp		hf		hm	ht		hs		hn	hl	hr	hc		hx	hk

Key

Color	Meaning
Yellow	Level II heterosyllabic clusters
Light green	Level III heterosyllabic clusters that are also Level I coda clusters
Light blue	Other Level III heterosyllabic clusters
Indigo	Level III onset clusters (may occur intervocalically)

Figure 3: Word-medial clusters of Level I [WORK IN PROGRESS]

	p	b	f	v	m	t	d	s	z	n	l	r	c	j	x	k	g	h	q
p			pf		pm	pt	pd	ps	pz	pn	pl	pr	pc	pj	px	pk	pg	ph
b				bv	bm	bt	bd	bs	bz	bn	bl	br	bc	bj	bx	bk	bg	bh
f	fp				fm	ft	fd	fs	fz	fn	fl	fr	fc	fj	fx	fk	fg	fh
v		vb			vm	vt	vd	vs	vz	vn	vl	vr	vc	vj	vx	vk	vg	vh
m	mp	mb	mf	mv		mt	md	ms	mz	mn	ml	(mr)	mc	mj	mx	mk	mg	mh	mq
t	tp	tb	tf	tv	tm			ts		tn	tl	tr				tk	tg	th
d	dp	db	df	dv	dm				dz	dn	dl	dr				dk	dg	dh
s	sp	sb	sf	sv	sm	st				sn	sl	sr	(sc)	(sj)	(sx)	sk	(sg)	(sh)
z	zp	zb	zf	zv	zm		zd			zn	zl	zr	(zc)	(zj)	(zx)	zk	zg	zh
n	np	nb	nf	nv	nm	nt	nd	ns	nz		nl	(nr)	nc	nj	(nx)	nk	ng	nh	nq
l	lp	lb	lf	lv	lm	lt	ld	ls	lz	ln		(lr)	lc	lj	lx	lk	lg	lh	lq
r	rp	rb	rf	rv	rm	rt	rd	rs	rz	rn	rl		rc	rj	rx	rk	rg	rh	rq
c	cp	cb	cf	cv	cm	ct	cd	(cs)	(cz)	cn	cl	(cr)				ck	cg	ch
j	jp	jb	jf	jv	jm	jt	jd	(js)	(jz)	jn	jl	jr				jk	jg	jh
x	xp	xb	xf	xv	xm	xt	xd	(xs)	(xz)	xn	xl	xr	xc			xk	xg	xh
k	kp	kb	kf	kv	km	kt	kd	ks	kz	kn	kl	kr	kc	kj	kx			kh
g	gp	gb	gf	gv	gm	gt	gd	gs	gz	gn	gl	gr	gc	gj	gx
h	hp	hb	hf	hv	hm	ht	hd	hs	hz	hn	hl	hr	hc	hj	hx	hk
q	qp	qb	qf	qv	qm	qt	qd	qs	qz	qn	ql	qr	qc	qj	qx	qk	qg	qh

Key

Color	Meaning
Pink	Level I heterosyllabic clusters
Yellow	Level II heterosyllabic clusters
Light green	Level III heterosyllabic clusters that are also Level I coda clusters
Light blue	Other Level III heterosyllabic clusters
Indigo	Level III onset clusters (may occur intervocalically)