CHAMBERS'S INFORMATION FOR THE PEOPLE. 



deduced from them for forming the past tense of 

 a new verb. Besides, reduplication is only suited 

 to simple verbs root or stem verbs. Accordingly, 

 in the case of compound and derivative verbs, the 

 expedient was adopted of affixing an auxiliary 

 verb. In Latin, as already observed, a common 

 form was vi or ui (ama-vi, par-ui), corrupted from 

 fuii which is a contraction of the reduplicated 

 past fu/uf, of fit, to be. In perfects like scrip-si, 

 we may recognise the relics of an obsolete redupli- 

 cate perfect of ts-, to be. The Teutonic languages 

 employed the reduplicated preterit of the verb 

 'do.' That the modern signs of past time in 

 English (d or ed\ and in German (te), are relics 

 of Ang.-Sax. dide, and Old German Ma, is rendered 

 certain by the occurrence of such forms in Moeso- 

 Gothic as salbo-ded-um, tame-ded-um, which are 

 as explicit as 'anoint-did-we,' 'tame-did-we.' The 

 preterits formed in this way soon came to out- 

 number the older reduplicated forms, and we now 

 speak of them as ' regular,' and of the others as 

 'irregular.' Many verbs that in Old English 

 formed the past tense by a change of vowel, 

 now follow the prevailing fashion, and take d 

 as leap, which had at one time lap in the 

 past tense, and has so still in some provincial 

 dialects. 



Assimilation. When the composition of words 

 brings incompatible letters together, one of the two 

 is assimilated to the other. The same takes place 

 when the falling-out of a vowel allows two incom- 

 patible articulations to come together. The opera- 

 tion of this law is conspicuously seen in words from 

 Latin and Greek, compounded with prepositions 

 and other prefixes ; as <z/-firm for ad-firm, ar-rogate 

 for <z</-rogate, //-licit for in-licit, sup-port for sup- 

 port, sym-metry for syn-metry. In these cases, 

 the assimilation is complete ; but the same end, 

 namely, ease of pronunciation, is often attained 

 by approximation merely, as *#/-pute for /-pute, 

 in which the transition to the labial p is easier 

 from the labial nasal m, than it would have been 

 from the lingual nasal n. The case of prefixes, 

 however, is of less importance in comparative 

 philology than other parts of the process of word- 

 building. Whenever we form a new compound, 

 although we may retain the spelling, we instinc- 

 tively accommodate the pronunciation of one 

 part to that of the other. Thus, handkerchief is 

 pronounced hang-kerchief. Here d is altogether 

 dropped, and the lingual is changed to the 

 guttural ng, in order to make the transition to k \ 

 more easy. 



In the earlier and more fluent stages of lan- 

 guage, this process of accommodation was more 

 rapid and complete, being without the drag put 

 upon it by writing and printing. Of complete 

 assimilation of the preceding consonant to the 

 following, we have examples in the Latin words 

 sum-mus for sup-mus, sel-la for sed-la (sed-, to sit), 

 puel-la for puer-la, gram-ma for graph-ma. In 

 longis-simus for longis-timus,facil-limus for facil- \ 

 limits, the latter is assimilated to the former. 

 Approximation is seen in scri/-tus for scri^-tus, 

 ar-tus for a^-tus (ag-, to do or drive), sowz-nus for 

 so/-nus (compare soporific), Sawz-nium for Sa- 

 nium. Instead of assimilation, one of the conson- j 

 nants is frequently dropped, in which case the 

 vowel is usually lengthened by way of compensa- 

 tion. For example, the Latin for foot was origin- 

 ally ped-s, which first, probably, became pess, and 



24 



was then written pis; lu-na is for luc-na (luc-, to 

 shine), lu-men for luc-men. 



Sometimes ease of pronunciation is sought in 

 dissimilation: as in equej-ter for eque/-ter, clauj- 

 trum for claurf-trum. A great many adjectives in 

 Latin were formed by the affix -a/is, as mortalis, 

 regalis ; but when the stem contained / in the last 

 syllable, -arts was substituted : e.g.popul-aris, vulg- 

 aris. In caeru-leus for cas/u-leus (compare ccelum, 

 heaven), it is the stem that is changed. 



In making these accommodations each language 

 has peculiarities of its own. The Latins, for ex- 

 ample, found such a combination as genesis (the 

 original form of the genitive of genus) inconven- 

 ient, and changed the first s into r generis; the 

 Greeks threw it out altogether, and made genesos 

 into geneos, and then into genous. The ancient 

 Celtic inhabitants of Spain disliked initial sp, st, as 

 their neighbours of Gaul did (see p. 22), and changed 

 Latin spata into espada, stabilis into estable. 

 Modern Italian tolerates sp, st, but not fl, pi, and 

 turns Latin fior into fiore, planus into piano. 

 French and Spanish retain these combinations, 

 havingy&#r, fior, and plain, piano. Spanish has 

 the peculiarity of turning Latin initial /, in many 

 cases, into h, so that, e.g. facere, filius, fames, 

 formosus, appear disguised as hacer, hijo, hambre, 

 hermoso. 



These examples will suffice to shew how, what 

 were originally the same words, may, in the 

 mouths of people living apart, and with different 

 idiosyncrasies, after a lapse of years, become 

 so different as to form a new language. This 

 leads us to the subject of 



DIALECTS. 



In speaking of a people having essentially all 

 one language, but living extended over a wide 

 territory, the name of dialects is given to those 

 varieties or peculiar forms which that language 

 assumes among the various tribes or other local 

 divisions of the people. It is clear that the wider 

 the separation comes to be between the several 

 tribes, and the more they differ in mode of life and 

 other circumstances, the more marked will the 

 differences of dialect become. When, again, a par- 

 ticular tribe of this people increases in numbers, 

 and also extends its territory, the same process is 

 repeated, and its dialect becomes broken into a 

 number of sub-dialects. The principal check to 

 this tendency to seemingly endless subdivision of 

 language, is furnished by an increasing degree of 

 common culture and civilisation. Where this is 

 wanting, as in Africa and among the native popu- 

 lations of America, the subdivision is practically 

 endless. 



A further check to divergence is usually found 

 in one dialect of a country acquiring a dominance 

 over the others. Various circumstances may give 

 rise to this. The tribe that succeeds in establish- 

 ing political supremacy is likely to make its dialect 

 also prevail. A popular poet also may give his 

 local dialect a general currency. Accidental cir- 

 cumstances have, in many cases, decided the 

 rivalry. The Bible happened to be translated by 

 a High-German, Luther, into his native dialect ; 

 other works on the then all-engrossing subject of 

 religion followed in the same dialect ; happily, too, 

 the art of printing had just attained the perfec- 

 tion necessary to give these productions general 



