1. _The Libyan Dialects_ (mostly misnamed "Berber languages," after an
unfortunate, vague Arabic designation, _barabra_, "people of foreign
language"). The representatives of this large group extend from the
Senegal river (where they are called Zenaga; imperfect _Grammaire_ by
L. Faidherbe, 1877) and from Timbuktu (dialect of the Auelimmiden,
sketched by Heinrich Barth, _Travels_, vol. v., 1857) to the oases of
Aujila (Bengazi) and of Siwa on the western border of Egypt.
Consequently, these "dialects" differ more strongly from each other
than, e.g. the Semitic languages do between themselves. The purest
representative seems to be the language of the Algerian mountaineers
(Kabyles), especially that of the Zuawa (Zouaves) tribe, described by
A. Hanoteau, _Essai de grammaire kabyle_ (1858); Ben Sedira, _Cours de
langue kab._ (1887); _Dictionnaire_ by Olivier (1878). The learned
little _Manuel de langue kabyle_, by R. Basset (1887) is an
introduction to the study of the many dialects with full bibliography,
cf. also Basset's _Notes de lexicographie berbere_ (1883 foll.). (The
dictionaries by Brosselard and Venture de Paradis are imperfect.) The
best now described is Shilh(a). a Moroccan dialect (H. Stumme,
_Handbuch des Schilhischen_, 1899), but it is an inferior dialect.
That of Ghat in Tripoli underlies the _Grammar_ of F. W. Newman (1845)
and the _Grammaire Tamashek_ of Hanoteau (1860); cf. also the
_Dictionnaire_ of Cid Kaoui (1900). Neither medieval reports on the
language spoken by the Guanches of the Canary Islands (fullest in A.
Berthelot, _Antiquites canariennes_, 1879; akin to Shilha; by no means
primitive Libyan untouched by Arabic), nor the modern dialect of Siwa
(still little known; tentative grammar by Basset, 1890), have
justified hopes of finding a pure Libyan dialect. Of a few literary
attempts in Arabic letters the religious _Poeme de Cabi_ (ed. Basset,
_Journ. asiatique_, vii. 476) is the most remarkable. The imperfect
native writing (named _tifinaghen_), a derivation from the Sabaean
alphabet (not, as Halevy claimed, from the Punic), still in use among
the Sahara tribes, can be traced to the 2nd century B.C. (bilingual
inscription of Tucca, &c.; cf. J. Halevy, _Essai d'epigraphie
libyque_, 1875), but hardly ever served for literary uses.
2. _The Cushitic or Ethiopian Family._--The nearest relative of Libyan
is not Ancient Egyptian but the language of the nomadic Bisharin or
Beja of the Nubian Desert (cf. H. Almkvist, _Die Bischari Sprache_,
1881 [the northern dialect], and L. Reinisch, _Die Bedauye Sprache_,
1893, _Worterbuch_, 1895). The speech of the peoples occupying the
lowland east of Abyssinia, the Saho (Reinisch, grammar in _Zeitschrift
d. deutschen morgenland. Gesellschaft_, 32, 1878; _Texte_, 1889;
_Worterbuch_, 1890; cf. also Reinisch, Die Sprache der Irob Saho,
1878), and the Afar or Danakil (Reinisch, _Die Afar Sprache_, 1887; G.
Colizza, _Lingua Afar_, 1887), merely dialects of one language, form
the connecting link with the southern Hamitic group, i.e. Somali
(Reinisch, _Somali Sprache_, 1900-1903, 3 vols.; Larajasse und de
Sampont, _Practical Grammar of the Somali Language_, 1897; imperfect
sketches by Hunter, 1880, and Schleicher, 1890), and Galla (L.
Tutscheck, _Grammar_, 1845, _Lexicon_, 1844; Massaja, _Lectiones_,
1877; G. F. F. Praetorius, _Zur Grammatik der Gallasprache_, 1893,
&c.). All these Cushitic languages, extending from Egypt to the
equator, are separated by Reinisch as _Lower Cushitic_ from the _High
Cushitic_ group, i.e. the many dialects spoken by tribes dwelling in
the Abyssinian highlands or south of Abyssinia. Of the original
inhabitants of Abyssinia, called collectively Agau (or Agau) by the
Abyssinians, or Falashas (this name principally for Jewish tribes),
Reinisch considers the Bilin or Bogos tribe as preserving the most
archaic dialect (_Die Bilin Sprache_, Texts, 1883; _Grammatik_, 1882;
_Worterbuch_, 1887); the same scholar gave sketches of the Khamir
(1884) and Quara (1885) dialects. On other dialects, struggling
against the spreading Semitic tongues (Tigre, Amharic, &c.), see Conti
Rossini, "Appunti sulla lingua Khamta," in _Giorn. soc. orient._
(1905); Waldmeyer, _Wortersammlung_ (1868); J. Halevy, "Essai sur la
langue Agaou" (_Actes soc. philologique_, 1873), &c. Similar dialects
are those of the Sid(d)ama tribes, south of Abyssinia, of which only
Kaf(f)a (Reinisch, _Die Kafa Sprache_, 1888) is known at all fully. Of
the various other dialects (Kullo, Tambaro, &c.), vocabularies only
are known; cf. Borelli, _Ethiopie meridionale_ (1890). (On Hausa see
below.)
There is no question that the northernmost Hamitic languages have
preserved best the original wealth of inflections which reminds us so
strongly of the formal riches of southern Semitic. Libyan and Beja
are the best-preserved types, and the latter especially may be called
the Sanskrit of Hamitic. The other Cushitic tongues exhibit increasing
agglutinative tendencies the farther we go south, although single
archaisms are found even in Somali. The early isolated High Cushitic
tongues (originally branched off from a stock common with Galla and
Somali) diverge most strongly from the original type. Already the Agau
dialects are full of very peculiar developments; the Hamitic character
of the Sid(d)ama languages can be traced only by lengthy comparisons.
The simple and pretty Haus(s)a language, the commercial language of
the whole Niger region and beyond (Schoen, _Grammar_, 1862,
_Dictionary_, 1876; Charles H. Robinson, 1897, in Robinson and
Brookes's _Dictionary_) has fairly well preserved its Hamitic grammar,
though its vocabulary was much influenced by the surrounding Negro
languages. It is no relative of Libyan (though it has experienced some
Libyan influences), but comes from the (High ?) Cushitic family; its
exact place in this family remains to be determined. Various languages
of the Niger region _were_ once Hamitic like Haus(s)a, or at least
under some Hamitic influence, but have now lost that character too far
to be classified as Hamitic, e.g. the Muzuk or Musgu language (F.
Muller, 1886). The often-raised question of some (very remote)
relationship between Hamitic and the great Bantu family is still
undecided; more doubtful is that with the interesting Ful (a) language
in the western Sudan, but a relationship with the Nilotic branch of
negro languages is impossible (though a few of these, e.g. Nuba, have
borrowed some words from neighbouring Hamitic peoples). The
development of a grammatical gender, this principal characteristic of
Semito-Hamitic, in Bari and Masai, may be rather accidental than
borrowed; certainly, the same phenomenon in Hottentot does not justify
the attempt often made to classify this with Hamitic.
3. _Ancient Egyptian_, as we have seen, does not form the connecting
link between Libyan and Cushitic which its geographical position would
lead us to expect. It represents a third independent branch, or rather
a second one, Libyan and Cushitic forming one division of Hamitic. A
few resemblances with Libyan (M. de Rochemonteix in _Memoires du
congres internat. des orientalistes_, Paris, 1873; elementary) are
less due to original relationship than to the general better
preservation of the northern idioms (see above). Frequent attempts to
detach Egyptian from Hamitic and to attribute it to a Semitic
immigration later than that of the other Hamites cannot be proved.
Egyptian is, in many respects, more remote from Semitic than the
Libyan-Cushitic division, being more agglutinative than the better
types of its sister branch, having lost the most characteristic verbal
flection (the Hamito-Semitic imperfect), forming the nominal plural in
its own peculiar fashion, &c. The advantage of Egyptian, that it is
represented in texts of 3000 B.C., while the sister tongues exist only
in forms 5000 years later, allows us, e.g. to trace the Semitic
principle of triliteral roots more clearly in Egyptian; but still the
latter tongue is hardly more characteristically archaic or nearer
Semitic than Beja or Kabylic.
All this is said principally of the grammar. Of the vocabulary it must
not be forgotten that none of the Hamitic tongues remained untouched
by Semitic influences after the separation of the Hamites and Semites,
say 4000 or 6000 B.C. Repeated Semitic immigrations and influences
have brought so many layers of loan-words that it is questionable if
any modern Hamitic language has now more than 10% of original Hamitic
words. Which Semitic resemblances are due to original affinity, which
come from pre-Christian immigrations, which from later influences, are
difficult questions not yet faced by science; e.g. the half-Arabic
numerals of Libyan have often been quoted as a proof of primitive
Hamito-Semitic kinship, but they are probably only a gift of some Arab
invasion, prehistoric for us. Arab tribes seem to have repeatedly
swept over the whole area of the Hamites, long before the time of
Mahomet, and to have left deep impressions on races and languages, but
none of these migrations stands in the full light of history (not even
that of the Gee'z tribes of Abyssinia). Egyptian exhibits constant
influences from its Canaanitish neighbours; it is crammed with such
loan-words already in 3000 B.C.; new affluxes can be traced,
especially c. 1600. (The Punic influences on Libyan are, however, very
slight, inferior to the Latin.) Hence the relations of Semitic and
Hamitic still require many investigations in detail, for which the
works of Reinisch and Basset have merely built up a basis.
(W. M. M.)