in Anglo-Saxon than it is in Modern English. In a complete inflectional system the verb would always indicate by its form the concord in person with the subject. But Anglo-Saxon did not possess a complete system. The plural forms never indicated distinctions of person, and the first and third persons of the preterite were always alike. The fullest indication of person was in the present singular, as in the modern archaic I sing, thou singest, he singeth, corresponding to Anglo-Saxon ic singe, ðū singest, he singeth. Already in Anglo-Saxon times, however, a tendency appeared to level the personal endings of the present tense. This tendency was carried farthest in the northern dialect, where the ending s was generalized as an ending for all forms of the present, singular and plural. If this northern use had persisted into Modern English, our present tense would be I sings, you sings, he sings, we, you, they sings, the s becoming merely the inflection for the present, with no personal significance. In the Middle English period there was more or less dialectal variation in the forms of the present, and the northern custom has left one trace in Modern English in the third singular present in s. This is the only personal inflection left in the Modern English verb, and it serves no useful purpose. If it is possible to say I sang, you sang, he sang without danger of misunderstanding, it would be equally possible to say I sing, you sing, he sing. As other examples have shown, however, consistency was not a virtue that was carried to its limits in those periods of the language which determined its modern shape.

Number in the Anglo-Saxon verb followed a course similar to that of person. In general the Anglo-Saxon verb distinguished plurals from singulars by means of

distinctive endings. The endings for the most part were weak unstressed syllables, however, and therefore shared the general tendency of such syllables to be slighted in pronunciation and ultimately to disappear. Since number would be indicated anyway by the subject, and since the concord between the number inflection of the verb and the number of the subject was merely a matter of grammatical harmony, efficiency in expression on the side of meaning made no demand for the retention of number inflection in the verb. After a certain amount of wavering, the language settled upon a system which has only one surviving indication of number in the verb, the s ending of the present, which stands for three things, present tense, third person, and singular number.

The Anglo-Saxon strong verb customarily had a different vowel in the preterite plural from the vowel of the preterite singular. Thus the preterite singular of rīdan, to ride, is rād, but the preterite plural is ridon. None of these preterite plurals have survived as plurals, though when the preterite and the past participle of a strong verb were leveled under one form, sometimes the vowel of the plural survived as the characteristic vowel of the preterit tense, especially when the vowel of the preterite plural was the same as the vowel of the past participle. But the determining element in such developments was probably not the vowel of the preterite plural but the presence of the same vowel in the past participle.

Two moods besides the indicative were marked inflectionally in the Anglo-Saxon verb, the imperative and the optative or subjunctive. The imperative differed very slightly from the indicative in its inflectional forms, and in Modern English, it differs not at all from it. The

subjunctive had as its distinguishing marks those general utility endings, -e in the singular, -en in the plural. Neither of these was well fitted to resist the tendency towards the dropping of endings in the period of transition, and of course Modern English has no distinctive subjunctive endings. In the third singular present, the omission of the s of the indicative makes a subjunctive of it, corresponding to an Anglo-Saxon third singular present subjunctive with a final e. Even this is now an archaic use, and aside from occasional forms of the verb to be in conditions contrary to fact, the subjunctive meanings of older English are now expressed by a great variety of verb phrases.

Casting a backward glance over this rapid survey of some of the more striking instances of structural change that have taken place within the historical periods of English, the reader will observe that two main causes have been at the bottom of them The first cause was the tendency to obscure the values in pronunciation of inflectional syllables as they stood at the ends of words in unstressed position. As these syllables became obscured in sound, the grammatical values for which the syllables had formerly stood could less effectively be maintained. The consequence was that the syllables themselves tended to disappear and with them their grammatical significances. The thing that saved these grammatical endings from disappearing completely was that some of them, for example, the mark of the plural in nouns, had a logical significance which the language was bound to express in some way and which it therefore expressed by using a part of the traditional materials with which it was provided. The second cause was that drift towards regularity which moved in the direction of reduc

ing several mechanisms for doing the same linguistic labor, when several were present, to one. Thus nouns tended to become all of one type and verbs tended to follow the lead of the weak verb. But the tendency always stopped short of the complete realization of a perfectly regular simplified system. For it was a tendency of the practical world and of uncontrolled experience in which perfect systems rarely demand a place. It produced a working, not an ideal system in language.

XX

THE ENGLISH VOCABULARY

It is a simple and natural question to ask where the words of the English language ultimately came from, but not so easy to find a simple answer to this question. One thing that complicates the matter is that the conditions under which words come into being are no longer present in the English language and have not been present within historical periods. To invent words is now practically impossible, and yet all the words contained in the English language at some time or other must have been invented. But the period of invention is now closed, and during the whole civilized history of the English language the constructive linguistic process by which new words are made has been almost completely atrophied and inactive. Many words have been added to English within historical times by borrowing, but practically none by out and out invention. It is not impossible to make words that look, sound, and mean something like genuine words, but the psychological conditions which favor this kind of inventiveness and which lead to the adding of such new words to the language in practical use are no longer actively present. Such invented words are really not words until they pass current in general use, and the society in which the English language now exists is no longer willing to receive such new-comers into its circle. The invention of words is a genuinely primitive process in the growth of language. Perhaps one may say that