have been the more elegant, the expression is not ungrammatical. The sun shines to-day brighter than it did yesterday, and to-morrow it will shine brightest.-Here also the sense is adverbial. In words like oftener and seldomer the adverbial comparison is beyond doubt. § 335. Adverbs, then, take the degrees of comparison and not only do they do this, but the history of their forms is important. In Anglo-Saxon there were two forms; one in re and -este, the other in -or and -ost. Now the first of these was the form taken by adjectives; as se scearpre sweord the sharper sword, and se scearpeste sweord the sharpest sword: the second, the form taken by adverbs; as, se sweord scyrd scearpor=the sword cuts sharper, and se sweord scyrð scearpost = the sword cuts sharpest. § 336. More than this—the adverbial form had a tendency to make the preceding vowel full: the adjectival, a tendency to make it small. Thus— Be Of this change, the word last quoted is a still-existing specimen, as old, elder, and older, eldest and oldest. tween the two forms, however, there is a difference in meaning, elder being used as a substantive, and having a plural form, elders. This, however, is by the way. A more important word is rather: in which we pronounce the a like the a in father, or full. Nevertheless, the positive form is small, the a being pronounced as the a in fate, or small. The word itself means quick, easy= the classical root pad- in padios. What we do quickly and willingly we do by preference. If the word rather were an adjective, the vowel of the comparative would be sounded as the a in fate. As it is, however, it is adverbial, and as such is properly sounded full. CHAPTER XIII. ON THE ORDINALS.-WHAT IS THE CONNEXION BETWEEN THE ORDINALS AND THE DEGREES OF COMPARISON. § 337. THE Ordinals are derived from the Cardinals. There is, however, no etymological connection between either one and first, or two and second. With the others the ordinal form is either th or a modification of it. Thus § 338. Is there any connection between the Ordinals of Numerals and the Superlatives of Adjectives? It is an undoubted fact that more than one form is common to certain Superlatives, and to certain Ordinals. Thus the -m- in for-m-er, of which the Anglo-Saxon is for-m-a, and which is, in Latin, pri-m-us, and, in Lithuanic, pir-m-as, is, without doubt, the -m- in infi-m-us, exti-m-us, &c.= lowest, outermost, &c.; all being superlatives. It is also an undoubted fact that the -t- in sex-t-us (sixth) is the -t- in pw-T-os, and the -tim- in sep-tim-us, the tim of ex-tim-us. It is impossible to see these coincidences without admitting the possibility of such identifications. Those, however, who see this are asked to see more. They are asked to see in the Greek form -727-, and I-TAT-05 an original -raμT- in which both the -7and -μ- once existed. They are then asked to see, in a word like πęw-т-os, a form in which -μ- is lost, but the preserved. They are then asked to see in infi-mus, a form where the -t- is preserved, but without the -μ-. -7 § 339. All this passes within the region of the Superlative Degree, and without any hypothesis as to the affinity between the ideas of Superlativity and Ordinality. But what if the latter be superadded? In this case, the Ordinals are dealt with as Superlatives, and, mutatis mutandis, the reasoning is repeated. The -tim- in sep-timus is the full, perfect, and typical form; the -t- in quar-tus, the -t- minus m-. The -m- in deci-m-us is the -mminus t-: all this within the compass of one language. But this is not all; the Latin for 7 is septem, the Greek Inτa. The Norse for 7 is sjau. But, in the English, in seve-n, the -n- (being the -m- of the ordinal) is reflected back (so to say) on the cardinal. This may, or may not, be the case. But there is more behind. The Greek for 10 is dena; wherein, not only the -tbut the -m- is lost also; as may be seen from dec-em. But the English for 10 is ten; in Moso-Gothic taihun. Here the -h-=-k- (in dɛna), and -c- (in decem); whilst the -n- = -m- in septi-m-us = -m- in infi-m-us =-min pri-m-us-m- of the Superlative Degree-m- of ordinality-this-m- of ordinality being reflected on the Superlative. The same applies to seven and nine. The -n- is not radical, as is inferred from sjau, and ivvɛa : and it is ordinal, as is inferred from septi-m-us and novi-m-us=nonus. All this should be known, because it is found in the writings of authoritative grammarians. But is it true? I cannot say. It explains so much that I am slow to believe it wholly wrong. At the same time the patent and ostensible argument in favour of it is unsatisfactory. To treat first as the ordinal of two, is like treating I as the nominative of me. They are not only two words but the names for two different ideas. First is a superlative all the world over. It is at the most honourable end of a series, or order; and, as such, Ordinal. But this order, in which it is so superlative, is not represented by one, but by second, third, fourth, and so on. In respect to these it is both ordinal and superlative. What it is to one is another matter. It is certainly not its superlative. § 340. To proceed. Compare second with two, and what is the correlation? None. The true correlative to second is first; and as second is from the Latin secundus, to which the root is the sec- in seq-uor, the two together mean, there or thereabouts, the preceding and the following. If any degree of comparison comes in here, it is the comparative; and that this does come in is shown in those languages which, like the Danish, use anden other for second. = § 341. Notwithstanding all this, it is possible that, in words like third, fourth, &c., some idea of superlativeness may exist, though not to the extent to which it exists in first. When we say the fifth, or the sixth, we use the definite article just as we do when we say the best, or the worst. We also imply that a number of objects is spoken about; inasmuch as the fifth implies the fourth, third, second and first which preceded it-the highest number being at the head of the series. In this there are the elements of ordinality of some kind. But is it the ordi T nality that implies a cardinality? Is it a correlation between fifth and five? No. The ordinals, from two, upwards, are ordinal to each other, and not to their so called cardinals. CHAPTER XIV. EXPRESSION OF DIFFERENCE OF SEX.-DERIVATIVES IN -IN AND -STER. § 342. Forms in -in.-The chief affix by which the name of a male is converted into that of a female, is, in German-in; so that from freund=friend we get freund-inn =female-friend. It is a termination which is not only German but Sarmatian also: the Lithuanic giving = = This being the case, its absence in English is remarkable. The only word in which it is believed to exist at the present moment is vixen female fox = füchsinn, German. I am, however, by no means certain that the word is not of recent introduction. If so, it is in the same predicament as margravine and landgravine from marchgrave, and is merely a naturalized German word. That the -ine in hero-ine, from hero, has a wholly different origin is manifest; being from the Greek ngwéin. § 343. Forms in -ster.-Forms in -ster were originally the names of Females. The old glossaries give us Textor webba (1.) Citharedus hearpere |