Attempts were made to connect them with the Indo-Europeans; I think unsuccessfully. Of course there was a certain amount of relationship of some kind; but it by no means followed that this established the real affiliations. There was a connexion; but not the connexion. The reasons for this view lay partly in certain undoubted affinities with the Persians, and partly in the fact of the Jew, Syrian and Arab skulls, and the Jew, Syrian and Arab civilizations coming under the category of Caucasian.

Consciously or unconsciously, most writers have gone on this hypothesis—naturally, but inconsiderately. Hence the rough current opinion has been, that if the Semitic tribes were in any traceable degree of relationship with the other families of the earth, that relationship must be sought for amongst the Indo-Europeans.

The next step was to raise the Semitic class to the rank of a standard or measure for the affinities of unplaced families; and writers who investigated particular languages more readily inquired whether such languages were Semitic, than what the Semitic tongues were themselves. Unless I mistake the spirit in which many admirable investigations have been conducted, this led to the term Sub-Semitic. Men asked about the amount of Semitism in certain families as if it were a substantive and inherent property, rather than what Semitism itself consisted in.

And now Sub-Semitic tongues multiplied; since Sub-Semitism was a respectable thing to predicate of the object of one’s attention.

The ancient Ægyptian was stated to be Sub-Semitic—Benfey and others having done good work in making it so.

Mr. Newman did the same with the Berber. Meanwhile the anatomists acted much like the philologists, and brought the skulls of the old Ægyptians in the same class with those of the Jews and Arabs, so as to be Caucasian.

But the Caucasians had been put in a sort of antithesis to the Negros; and hence came mischief. Whatever may be the views of those able writers who have investigated the Sub-Semitic Africans, when pressed for definitions, it is not too much to say that, in practice, they have all acted as if the moment a class became Semitic, it ceased to be African. They have all looked one way; that being the way in which good Jews and Mahometans look—towards Mecca and Jerusalem. They have forgotten the phænomena of correlation. If Cæsar is like Pompey, Pompey must be like Cæsar. If African languages approach the Hebrew, the Hebrew must approach them. The attraction is mutual; and it is by no means a case of Mahomet and the mountain.

I believe that the Semitic elements of the Berber, the Coptic and the Galla are clear and unequivocal; in other words, that these languages are truly Sub-Semitic.

In the languages of Abyssinia, the Gheez and Tigré, admitted, as long as they have been known at all, to be Semitic, graduate through the Amharic, the Falasha, the Harargi, the Gafat, and other languages which may be well studied in Dr. Beke’s valuable comparative tables[20], into the Agow tongue, unequivocally indigenous to Abyssinia; and through this into the true Negro classes.