The etymological database for the Peripheral Khoisan (= Peripheral Bushman; = North-South Khoisan) language family, linked directly to the "Macro-Khoisan" database (higher level) and leading to subordinate databases for North Khoisan, ǂHoan, !Wi and Taa (lower levels).

Peripheral Khoisan is a not-universally accepted language grouping within Khoisan, suggested by G. Starostin in several publications, all of which are also available as text files in the "Articles and Books" section of the Tower of Babel site. Final demonstration for language relationship comes in the form of a complete reconstruction, with an exhaustive set of phonetic correspondences and a large etymological dictionary. The current database is still a preliminary approximation to this format: not all the correspondences have been established, and its current scope of around 1,000 etymologies (many of which are quite questionable) is not sufficient for a full-fledged etymological dictionary. Nevertheless, many correspondences, enough to conduct a relatively strict lexicostatistical survey, have been suggested, and the evidence for relationship is sufficient to consider further work on this subgrouping quite promising. Unlike the "Macro-Khoisan" database, which does not aspire to be anything else than mere comparanda for future consideration, Peripheral Khoisan is very much a reality, and work on it should probably be taken seriously.

Consonant and vowel correspondences between the main branches of Peripheral Khoisan - North Khoisan and South Khoisan - are quite complex, involving frequent click bifurcation and various non-trivial developments in the click efflux system. This is because, although North and South Khoisan per se consist of closely related languages or dialects, the two proto-languages have a much more ancient date of separation themselves (see the genealogical tree on the site). A very preliminary general draft looks as follows:

CLICK INFLUXES

Influx PNK ǂHoan PSK
*| *| | *|
*|1 *! ! *|
*ǂ ǂ
*ǂ1 *| |
**!̯ ! *!
*1 *!̯ ! *||
*! *! ! *!
*!1 *! ! *||
*|| *|| || *||
*||1 ǂ *||

Note: click mergers in NK and SK have led to, respectively, the statistical surge of the alveolar click (!) in NK and the lateral click (||) in SK (455 out of 2200 as opposed
to 330~350 in both Zhu and Nama).
The phenomenon of labial clicks in SK and ǂHoan is still not quite understood; most of them seem to have a secondary origin. Some of the roots involving labial reflexes in these subgroups are provisionally marked with a labialization sign (-ʷ-).

CLICK EFFLUXES

Efflux PNK ǂHoan PSK
*0 *0 0 *0
*k_ *g 0
*g *g g *g
*g_ *0 0 *g
*n *n n
ʔ *ʔ / *q[-Vh]
*h *ʔ~*h / *n[-Ṽ] *h / *0[-Vh]
*gh *gh *g
*nh *n̯ ʔn *n
*kh *kh h *h
*nkh *h / *n[-V̰] *n̯
*ḳ *kx
*x *x / *[||-]kx *x
*nx *n̯
x *ɣ / *0[-Vh]
*ɣ_ *x
*nɣ *nG
*kx *kx kx *kx / *0[-V̰]
*k_x *gx *kx
*gx *gx *gx
*q *0 q / 0[-ʔ] *q
*q_ *g *q
*nq *n *0
*qh *kh qh~h *qh / *[||1-]0 / *0[-V̰]
*q_h *h~*gh qh *qh
*nqh *n̯ q
*G (*gq) *g / *n̯[-N] G *G
*Gh *kh qh *Gh
*G_ *0 q *G
*nG *n̯~*n G *(n)G
*q̇ *ʔ / *n[-VNV] *qʔ / *0[-Vh]
*q̇_ *g *qʔ
*q̇h *qh
*x qh~x *qh
*gx *G(h)
*ʁ_ *kx *G(h)
*n *n / *0[-Vh] ʔn / h[-Vh] *ʔn
*gn *g *ʔn
*n̯ *n̯ 0~h *0

Note: underlined effluxes (as well as non-click consonants in the next section) indicate fluctuation between voiced and voiceless reflexes in NK and SK - a phenomenon that is quite frequent but so far, lacks a reliable explanation (most probably related to tonal features).

CONSONANTS

Phoneme PNK ǂHoan PSK
*b *b *0
*ʔm *m *ʔm
*t *t *t
*t_ *d ʒ́ *t
*th *th ćh *th
*tʔ *t *0
*thʔ *th ~ *t *h
*d *d ʒ́ *d
*d_ *t *d
*dx *dx ćx
*d_x *tx *dx
*tx *tx *tx
*tkx ćkx *tkx
*t́ *c c *t
*t́ʔ *t t *čʔ / *č[-V̰]
*d́ *z ʒ́ *d
*n *n n *n
*ʔn *d *ʔn
*dn *d~*dh ʔ|n *ʔ|n
*s *s c *s
*z *z z *s
*c *c *s / *c[-a]
*cʔ *cʔ *|q / *|n[-Ṽ]
*ch ch
*chʔ *cʔ ch *|q[+V̰h]
*z z
*ʒh *sh *ʒh
*ʒhʔ *ʒhʔ ž
*ʒ_x *cx *ʒx
*cx *cx *cx
*ckχ *ǂkx *ckx
*c_kχ *ɡǂx *ckx
*ʒ_gχ *ǂkx *ʒgx
č / *š[-i] *s
*č_ʔ *ǯʔ čʔ *cʔ
*čh *čh čh *ch
z
*ǯhʔ *ǯhʔ ch *ʒh
š *s
*šh *ch
ž *t
*ɡ!̯ ž *!qh
*ƛ_ *ɡ!̯~*!̯ š
*k_ *g *k
*kh *kh kh *kh
*g *g g *g
*x *x *x
*g *g [Taa] / *0 [!Wi]
*kx *kx *kx
*kx *čh- ~ *sh- ~ *čʔ- č (k?) *kx
*q *k q *q
*q_ *g *q
*qh *x *qh
*G *x x *G
*h *h *h

SIMPLE VOWELS AND DIPHTHONGS

Phoneme PNK ǂHoan PSK
*a *a a / [ʘ-]oa *a
*e *e / *[click-]ai a *e
*e *a
*i *i i
*o *o / *u[-Ri] o *o / *[P-]u
*o o *a
*u *u / *o[-ʔa] u~o *u
*ae *ae e
*ai *ai (a)i / [C-]ui *ai
*am *am am *am
*an *ã *an
*ao *ao au / [C-]eo *ao
*au *au au / [C-]iu *ao
*ei *i *ai
*eu *u o *au
*oa *oa oa *oa
*oe *oe *oe
*oi *ui oe
*om *om
*om̥ *om_ *um
*ou ou *ou
*ɔa *oa / *[Q-]o̰a *a
*ɔi *oi *ai
*ua *oa *ua
*ui *ui ui *ui
*um *om *um
*ŋ̥ *ũ ~ *ŋ a *a
*n̥ *ŋ ~ *ŋ̃ i ~ e *an

NASALIZED VOWELS AND DIPHTHONGS

Phoneme PNK ǂHoan PSK
*ã *ã *ã
*ɛ̃ *ã ~ *ẽĩ
*õ *ũ *õ
*ãĩ *ãĩ *ãĩ

PHARYNGEALIZED VOWELS AND DIPHTHONGS

Phoneme PNK ǂHoan PSK
*a̰ *a / *[Q-, ž-]a̰ *a̰
*ḛ *ai (a)i *a̰
*o̰ *o / *u[-u] / *[Q-]o̰ o̰ / [Q-]o *o̰ / *[qʔ-]o
*ɔ̰ *a̰
*ṵ *u *ṵ
*a̰e *a̰e
*a̰i *a̰i *a̰i
*a̰m *a̰m a̰m
*a̰n *a̰n
*o̰i *ui o̰e *ṵje
*o̰m *o̰m
*o̰m_ *om_ *o̰m
*ɔ̰a *[Q-]o̰a *a̰
*ṵe *o̰e *ṵV
*ŋ̰ *ŋ̰ *ã̰

BREATHY VOWELS AND DIPHTHONGS

Phoneme PNK ǂHoan PSK
*ah *ah / *a *ah
*eh *ai *eh
*oh *o / *u o *oh
h *o o *ah
*aih *ai ai *aih
*aoh *aoh ao

PHARYNGEALIZED BREATHY VOWELS AND DIPHTHONGS

Phoneme PNK ǂHoan PSK
*a̰h *a̰ / *a[-Vh] a̰~a / [ʘ-]oa h
*o̰h *o̰
*ɔ̰h *o̰ / *uh[-ŋ] *a̰h
*o̰uh *u *o̰uh
*ṵeh *o̰e ṵe
*ŋ̥h *ŋ̥, *ŋh *ãh, *a̰hn

COMPLEX VOWELS

Vowel PNK ǂHoan PSK
*aʔ *a / *[Q-]a̰ʔ a(ʔ) *aʔ ~ *a
*a̰ʔ *a̰ʔ *a
*oʔ *o *oʔ ~ *uʔ
*ɔʔ *oʔ oʔo *a
*uʔ *o *uʔ
*o̰ʔ *o̰ʔ *oʔ
*ohʔ *o(ʔ) oʔo *ohʔ
*a̰hʔ *a̰(ʔ) ~ *aʔa *a̰ʔh / *[Q-]ahʔ
*o̰hʔ *o̰(ʔ) o̰ʔo *ṵh ~ *oh
*ṵhʔ *uʔu *ṵh
*aʔa *aʔa [ʘ-]oa *aʔ / *a
*aʔi *aʔi ae *ae
*aʔu *aʔu *aʔu
*a̰ʔa *a̰ʔa a / [C-]i *a / *aʔ / *a̰
*a̰ʔe *aʔe / *a̰ʔi *a̰e
*a̰ʔo *a̰ʔo *ao
*a̰ʔu *a̰ʔu *a̰u
*oʔa *oa *oʔa
*oʔo *oʔo *uʔ
*o̰ʔo *o̰ʔo *o ~ *o̰
*ɔʔa *oʔa *a
*ɔ̰ʔi *o̰ʔi *aʔi
*uʔu *uʔu *u
*ṵʔu *o̰ʔo *ṵ

For many more details on these and other correspondences as well as suggestions on eventual improvements, please see G. Starostin's article "From Modern Khoisan Languages to Proto-Khoisan" in the "Articles and Books" section of the site. The database will be subject to further elaboration based on those and newer ideas.

The database consists of the following fields:
1. Proto-Peripheral Khoisan (Proto-Bushman for short): the provisionary protoform for PPK.
2. Meaning: the general meaning(s) of the reconstructed stem.
3. Khoisan etymology: link to the "Macro-Khoisan" database where available.
4. Proto-Zhu: reflexation in the North Khoisan (Zhu) daughter branch, linked to the North Khoisan database.
5. Proto-!Wi: reflexation in the !Wi subbranch of Southern Khoisan, linked to the !Wi database.
6. Proto-Taa: reflexation in the Taa subbranch of Southern Khoisan, linked to the Taa database.
7. ǂHoan: reflexation in the ǂHoan language, linked to the ǂHoan database.
8. Notes: additional comments and considerations.
9. References: bibliographical links (mostly just references to previous suggestions of the same etymologies; references to language data sources should be found in the subordinate databases for North Khoisan, !Wi, Taa, and ǂHoan).

Notes on transcription of Proto-Peripheral Khoisan entries:
Clicks: | = dental click, ǂ = palatal click, ! = alveolar click, || = lateral click. The sign $ denotes an unidentified click influx (this only happens where all the data are drawn from old sources like Bleek 1956 and the evidence is highly ambiguous).
Click effluxes include zero (no special marking), voiced articulation (ɡ|, etc.), nasalisation (ɳ|, etc.), aspiration (|h, etc.), velar fricative (|x, etc.), velar affricate (|kx, etc.), uvular stop (|q, etc.), glottal stop (|ʔ, etc.). There are also a few rather arbitrary notations denoting specific correspondences for which adequate phonetic hypotheses have not yet been offered, including, e. g., -χ- and -ʁ- (uvular fricative effluxes?).
Non-click consonants: c, ʒ = hissing voiceless and voiced affricates; č, ǯ = hushing voiceless and voiced affricates.
Vowels: ɛ, ɔ = open correlates to closed e, o. The tilde sign denotes nasalisation; pharyngealized vowels are marked as a̰, o̰, etc.; breathy vowels are marked with a superscript h.
For details on transcription of specific languages and intermediate reconstructions please refer to the corresponding databases.