LANGUAGE. 



u br .aid. that the y liable* which the Chmeu employ a* affixes, hare 

 an eri*iaal meaning of their own ; for example, that the particle com- 

 nttnlruwd. above id, to denote the genltival relation, a at time* 

 Ml? nT*n i -T-*- equivalent to the L*tfn **, in other word*, 

 with ilw MM* of "departure." But here we have only a proof Uut 

 tfc, tllii*M grammar 1* lew corrupted, purer, more simple, than 

 our own ; tad may further appeal to thi* instance in support of the 

 doctrine, that all affix** wvrr in origin themselves root*. It may alao 

 b* noted that the uniwlnti of the gcnitival tuffli, though common in 

 ordinary Chine**, occur* meet rarely (" f*t niemaU,* *ayi Endlicher) 

 in thr Mandarine dialect In truth, the further we trace a language back, 

 taw nor* certain are we to And the suffixes expreated, aa in nAnu>- 

 rm. which fa but a corruption of n,lan-rv>t. Thus in the 

 OerakHl mtmd-t-l,cl>i, the middle syllable may well be regarded an 

 rtrtually a geoitival suffix, a better theory surely than that which 

 regard* It a* a euphonic Inmtlnn 



Bat while w* wouM avtribe the fake theory of the monosyllabic 

 character of Chine** to the fact of our approaching the language 

 thniecli it* syUabarium, *o we have the conrene error in the habit of 

 attributing a *p*daUy iwlysyllabic or polysynthetic character to Basque 

 and the language* of North America. Here there U, or rather was, no 



B language, and the traveller had no choice but to take down 

 what a native (poke, a* he ipoke it, and of course with many of what 



we luually regard a* (eparate word*, all jumbled together in one 

 confuMd ma**. W* need March for no better example than the poly- 

 *ylkbic term employed by the Axtecka in addreaaing a priest, as given 

 thi* word in the GEOGRAPHICAL DIVISION of this Cyclopaedia. 

 , *nd the inhabitant* of Iceland, uw for this purpose, the 

 r abort term, " Sir," but the Mexican layman addressed the 

 priest a* 'notlasmahuuttvpijccatdlzi*, which, being interpreted by 

 Clavigero, amounted to: no, "my," tlaamlli, " esteemed," inoAuufiY, 

 " revered." tto-pim>, " god-keeper, latii, " father." 



But the most formidable exception to thin auerted law of mono- 

 syllabic root* i* *een In the Semitic family, especially the Arabic 

 aad Hebrew, where it i* maintained that all ultimate word* are of 

 two *yllable*, and contain three consonants. Nay, some continental 

 scholar* of the present day, carry their theory of Hebrew root* t . ill- 

 extreme point of denying it the po**e*sion of any vowel, BO that li'. 

 B^ff 1 ''**; to them, U the root of the class of words signifying " kill." 

 Such theorist* seem to forget that spok< n language has a priority over 

 written language, for surely if they would picture to themselves the time 

 when Hebrew w** not yet written, tlit-y would find themselves in a" rc- 

 ductio ad absurdum," when declaring that to be a root which is con- 

 fessedly unpronounceable. No doubt in Hebrew, as we now have it, 

 the rowel* are very variable, so that for example, we find kotfl denotes 

 " killing," and kdiAl, " killed." The Greeks too have such varieties as 

 n/ut, pof , nifimu ; but thi* ha* not induced our classical scholars to 

 exhibit the root of thi* *erie* of word* a* rit. A* to the variation of 

 Towel* within the limit* of a root word, some remarks will presently 

 be made ; but to the doctrine that the root* of the Hebrew were 

 originally duyllabic, we may oppose the strong belief of some of the 

 ablest Hebrew scholars that this appearance is due solely to the fact, 

 that cecondary forms have supplanted the original simple stems. 

 Similarly in our language Mluw, reckon, listen, begin, tea/fair, have 

 pretty well usurped the place* of the obsolete verbs, Ml, red; Hit, yin, 

 Mtf ; *ad wuiiUrly in Latin, mna-rr, luna-re, lara-re, Knti-re, cone-re, 

 vci-ri, might have pawed for aimple verbs, had not tome of tin ..1,1, ,- 

 writers pi mil led lor us such furm a* ton-ere, tun-imui, tor-err, whilu 

 the iitmnr. at an early date of such verbal base* a* era- " puncture," 

 counting by prick of pin," ten- " feel,' men- " measure," i* esta- 

 Ui*h*d by derivative*, a* erm-iia, sU, cen-tnm, the Homeric t -t r-aa. ; 

 by the perf.am-n', *-(*, su.,K/i-nm, ten-la, *b.; and by menttu, manor, 

 menmra, *ad mrla (that in, men-ia), " a milestone," so to Bay. But 



i duyllabic ; yet the modern character of such a formation is at once 

 wuhltshiil by the consideration that the great body of thi* language 

 ha. demoded from the Sanskrit, in which the monosyllabic form of 

 primary verb* I* undisputed. 



The change of vowel In UOtand tdttl belong* to a question which 



2J mowconvenieotly considered presently, when the law of 

 QllMatQ YOWttal OOfiMal ttOflCT OOtaaHuCaTfttlOIL 



But it I* not uncommon for philologen to imagine to themselves a 

 di*Unction between what they are plowed to call n-nthetical and 

 analytical language* Thus they would contrast the habit of some 

 .Ury i.Us by affixes, as Mtri., ,(rcm, 



, , , , 



*r*>*f. tc., with utbcrs, eaneoially of modern date, 

 *** prefer to employ preposition* or formal --pronouns in the 



.I!???"*' W ^* fl ? 1 ? 7 J ?T" ; '"" il *"! "*> w the sequel that 

 there k. no ubsUntial distinction b*tw**n a caae-ending and* prepo- 

 aWon. that a.uiliary verb* are employed in thoM affixes whirl, Gelong 

 to pavfect tenwe, and that the final letter* of am*o, vribu, are a* 



eotupUUlypr.mu.in* in origin a* the / and yn> which an En K H*hn 

 pr^t, to hi. r,b. The reJ dktinction i*. thrt in the onToa.* 

 Itttlr word, precede, In the other, follow, unlew indeed any strew 

 la*! ntmo the corrupted form which oftm rUr*ni..n^-. .m.. 



thi 



" *" muow, uniew Indeed any straw be 

 form which often characterise, affixes. Yet 



nooii, raWt, being as fully pronounced a* our own by, the o of rt6o 

 carcely more abbreviated than our own 7, while I'm afraid, I 'it 

 done, I'll tell AIM, I'll lilt to l-noic, have cut down the verbs am, 

 kart, will, would, as far as possible short of annihilation. But take 

 such a case of contract as he did Inn compared with he lorect. Wln-n 

 we trace the Utter form upward, we find in the Anglo-Saxon he laved*; 

 and if we awend a step higher in the history of our language, so as to 

 enter the domain of the Gothic, the oldest specimen of the < i 

 family, we find good evidence that such a form as landed existed. l>ut 

 of course with the regular personal endings, which belong to the Gothic 

 |.r.-t. rite, attached. (Grimm's 'Or.,' i. 840.) Thus the contrast is 

 reduced to he lort-dtd a* opposed to he did Iwe ; in other words, we 

 have but a different arrangement of the same elements. 



In all cases, no doubt, were our means of inquiry equally abin 

 we should be brought to the conclusion that the format ; m . .1 . In I \ 

 is but the agglutination of significant syllable* to significant syllables, 

 or, in simpler language, of roots to root*. This assertion is foumU.l on 

 various considerations : first, the mere fact that in a large and rnpi<lly 

 increasing number of examples what were unintelligible affixes have 

 received their due explanation; secondly, the <l priori nrgi 

 already employed to show that roots obtained their significar 

 by arbitrary compact, but by natural association; and, thirdly, the 

 consideration that as affixes are attached to the end of words, and also 

 sustain for the time an inferior office, they are doubly liable to a 

 careless in other words, shortened pronunciation. On Kliip 

 for example, time is often of the first moment Thus, when the writer 

 once heard the French steersman of a French boat give orders to a 

 " mousse " for dropping the foresail, in the words "Down with her," 

 unconsciously speaking English, his inquiry as to the use ot 

 words was met by the remark that " Faites-le-descendre " would be too 

 long. No wonder, then, that sounds such as 6oe and fu.nl .-li.mlil I.e 

 preferred to ttoatneain and forecastle. Again, our own word tran'H, at 

 first sight of sufficiently obscure formation, has been analysed as repre- 

 senting wtn-tvic-cn, " ill brought-up," where wan is the prefix seen in 

 the old English van-hope, " mala spes," and still abundantly flouri^liin,L; 

 in the Dutch language ; while tuw-cn is the old participle of tli< 

 (oir, the equivalent of the German ;cA-rii and Latin </nr-ere; so that 



is substantially the same as the , "unedu- 



eAtcd, untrained." Our term is self-explained; but this 



wunl is beginning to obscure the second clement when pronounced as 

 leorkhtu. Again, take our word mum or ma'am who beforehand wmiM 

 have supposed that the final liquid in this word represented the Latin 



.,-iiiKt, as is the fact : nica domhia, madonna, madame, ma'am, 



ud in one of Dickens's novels, mi'm. 



Here we have seen the second port of the word shorn of its fair 

 proportions. The blow may however fall on the prefix. Thus, OKI, of 

 the Greek language, when used before verbs, was reduced iu the 

 .Kolian and Doric dialects to a* and op, and under special circum- 

 stances to a mere a or o. No wonder, then, that the Latin a-gnotc-erc 

 exhibits the same prefix as a mere vowel, or that the A. S. un-rnaw- 

 an, and Shaksperian a-cknov: (part, ackiniicn , ' Othello,' iit 8), .1 

 as our existing a-ckixnc-ledge, have taken .similar liberties. A still more 

 violent corruption of what might be proved to be the same pi ,.i. 

 Been in our e-lope, which represents the German oit-tauf-en , I 

 tmt4oop-4n. Cases oven occur where the original root is wholly 

 absorbed in some derived form. Thus, as Bopp has i 

 German im, corrupted from in-dc-m, has preserved the pre|wsiti> 

 the case-ending, but wholly lost that to which they were ti 

 attached, the element de (our //). Another example i .-imil.ir 

 extreme violence has been seen in /V for / ,/,/, the essential ~yl- 



/ of the verb having utterly disappeared. A* compression, 

 then, is a very common occurrence in language, it behoves an n 

 not hastily to assume the radical character of all monosyllabic verbs. 

 If he find two consonants either commencing or terminal 

 l.iUe, the strong probability is that a really disyllabic w. 

 him. Thiw. < c-k w but one consonant), ./,;), /,//, h. 



regarded as e of some such form as kru-tne or con 



,( or liiil-iH-l-, and so only secondary formations from th.- 

 simple km or r,,ti, /,//, coal, hull. Nay, in many cases win ! 

 initial consonant is a simple .-. /, or <i, a truncation of the < 

 letters may be suspected. For example, the I 

 cally known to be a corruption of ynote-trt, where the first F\ 

 repreeenU our kii-w, and thun yen or gon, corresponding to our 

 can, is the base. The rekted Greek anna has preserved tlie n 

 vowel, and only lost the guttural. So, again, when 



1 1 U placed beside the Greek y<\a-tiv, yt\<w-<Tu (yt\a^u. |io,ir 

 .<! see that the base of the words is gel. For an example of a 



il r, we may quote our verb run, which in the Dorset dialect is 



Aim, and here the n is but the remnant of a common suffix, as seen in 



a, 'I'h. IMHC, then, is Air, correspouding 



to the Latin rv, the interchange of h and c being the law 



between these language*. 



In the formation of secondary words, by such agglutinations a* we 

 have spoken of, due attention must be paid to those very natural laws 

 whii h n ,|iiiie an adaptation of discordant elements, a Kubject treated in 

 all it* integrity in Sanskrit grammar*, under t; com- 



position." The main principle* are the game for all languages. Thus, 

 when medial or thick consonant*, so to say, ore brought into contact 



