after the first radical. Dr. Oppert calls it the middle voice of Pael, strengthening the latter conjugation: thus in Kal halacu, "to go," Pael hallacu, "to make go," Iphtaal attallacu, "to be driven to go,' ""ambulare."

An instance of Iphtaneal is the aorist ultanpiru (for ustanappiru), as distinguished from the present ultanapparu.

As in Iphteal, verbs with i in Kal aorist may substitute te (ti) for a after the dental, e.g. yuptekid. The same takes place with verbs ; thus, lutebus, lutibbus.

Neither the Permansive nor the Imperative have been found.

Shaphel. This is one of the most commonly-used conjugations in Assyrian, and is formed by a prothetic s. Like the Aramaic Shaphel, presupposed in Arabic istaktala (conjugation 10) and Ethiopic ystagabbala, Hebrew and Phoenician Hiphil, Arabic and Ethiopic aktala, Aramaic and Assyrian Aphel, the conjugation has a factitive meaning. I would refer it to the root which appears in the Arabic shaha, wish," attached to the verb, like sa in Arabic (from saufa,

66

8

"in the end"), which is prefixed to the Imperfect to express futurity. As has become h in Hebrew, etc., it must have been initial, so that the peculiarly Hebrew root wy is excluded. A large number of roots in the various Semitic tongues, even in those which, like the Hebrew, have lost nearly every trace of Shaphel, are really Shaphel forms, e.g.

כון from שכן חר from שחר

In verbs y, a after the characteristics becomes e; e.g. usebis, useli. In the later inscriptions this change of consonant is sometimes transferred to the regular verbs, as in usescin, tuseznin, musecnis; just as a in Babylonian tends to become e (see p. 26); and as we get uptekid, etc., in Iphtaal and Iphteal.

The vowel of the characteristic may be dropped; e.g. usziz and even ulziz for usaziz.

For the Imperative see p. 58.

The Permansive has not been found. Dr. Hincks restores it as satcan.

phal, Arabic Tenth conjugation, Ethiopic ystagabbara, Hebrew Hithpael, and has a desiderative signification.

Verbs y' have e after the dental instead of a, e.g. ultebis. This is imitated by other verbs in the Babylonian period; e.g. ultesib and usteni'edu.

The Permansive Dr. Hincks believes would be satsecan.

After the example of Iphteal, another form of Istaphal, without the preformative u, seems to have come into use in the later period of the language. Thus we find in the Achæmenian inscriptions altabus (a corrupt form) by the side of ultebis, and istandhakhu may be another instance from Shalmaneser; but this is rather an Iphtaneal from

.

Aphel. This conjugation is confined, so far as I know, to the concave verbs, and will be treated of under them.

Itaphal.-Dr. Oppert quotes from the syllabaries itatspur as an example of this conjugation. The form ought to be yutatspir; itatspur will stand by the side of altabus above; but I should prefer to regard it as standing for the Ittaphal ittatspur.

Shaphael.-The same grammatical regularity that distin

guishes Assyrian among the Semitic languages like Sanskrit among the Aryan languages, producing the secondary conjugations with every voice, has also displayed itself in the Causative conjugation. Kal and Pael, answering to the aorist and present tenses, were regarded as the primary voices; to each of these was attached a causative in (u)sa. Each of the four forms thus obtained had a Passive assigned to it, the Reflexive Niphal being set apart for the Passive of Kal, as otherwise standing outside the regular verbal scheme -and finally all were provided with a secondary conjugation in t and tan. Shaphael is rarely found in the strong verb, as e.g. in yusnammir; but it frequently takes the place of Shaphel in verbs : thus usdhibbu‘, usmallu', usrabbi'. The Permansive may have had the form sasaccan; but it has not been found.

The Present is usnammar, the Aorist usnammir.

The vowel after s is regularly dropped on account of the weight of the following syllable.

The Imperative was probably susuccin. The Participle is

musnammiru.

Istaphael.-Here we find yusteni'edi for Aorist, ustamalta' for Present. The other tenses have not been detected.

The Passives.-I have already given my reasons for not considering forms like ilubusu as Passives of Kal, but as examples of a Poel.

As examples of the Passive of Pael, we have for the Permansive nuśśuku third plural masculine, nuśśuka third plural feminine, nummuru, summukhu, etc. In the Present we find yubullat, in the Aorist yubullit. Judging from Arabic analogy, there was no Imperative. I can add nothing to

what I have already said about the Passives of the remaining conjugations. The Passive of Shaphael ought to be sunummur or sunammur Permansive, yusnummar Present, and yusnummir Aorist. The Passive is never formed, as in Aramaic, by the dental. A solitary Aramaising form is itpisu for etpisu, "constituted," and here the dental is inserted after the first radical, while the word is only a nomen verbi. Traces of other conjugations, or rather nomina verbi, such as papel, pealpel, etc., will be found (see further on) under the head of the nomina verbi.

THE DEFECTIVE VERBS,

Verbs -These verbs follow the example of Niphal, assimilating the nasal when followed by any consonant except h or n, and the consonant is doubled. Before n and the vowels the first radical remains unchanged. Nn is never written n; thus we never find inamar for innamar, "it is

The Aorist of Kal takes u, a, and i after the second radical. Among those that have u are naʻamu, nabalu, nagagu, namaru, naśakhu, naśacu, napakhu, napaku, natsaru, nakabu, nakaru.

Among those that have i nadanu, nakhatsu, nadhu, nacalu, nacamu, nacaśu, nacaru, natsagu, nasagu, nasaku, nasaru. Among those that have a: na'aru, naharu, nazalu, nazaru, nakhazu, nadhalu, napalu, natsabu, natsatsu, nakamu.

The nasal is sometimes irregularly retained, more especially in the Achæmenian period. Thus we have indin for iddin, mandattu and mandantu for maddattu. It is possible, however, that the n was frequently not pronounced, though