The progressive reduction of the number of inflexional forms is a phenomenon not at all peculiar to English. On the contrary, most of the inflected languages of which the history is known have, to a greater or less extent, undergone the same kind of change. For example, although Modern High German is, as we have shown, much more complicated in its accidence than Modern English, it is much less so than the Old High German of a thousand years ago; the grammar of Old High German is simpler than that of Primitive Germanic, which was spoken at the beginning of the Christian era; and Primitive Germanic itself had retained only a comparatively small remnant of the profusion of inflexional forms possessed by the Primitive Indo-Germanic tongue, from which it was descended in common with Sanskrit, Greek, and Latin. We may note in passing that peasant German has lost much more of its original grammar than has the German spoken by educated people. This fact teaches us that culture is one of the influences which retard the process of simplification. But it should be remembered that culture may exist without books: there have been peoples in which there was little or no reading and writing, but in which nevertheless the arts of poetry and oratory were highly developed, and traditional correctness of speech was sedulously cultivated.
It is not wonderful that the tendency to simplification of accidence should be widely prevalent. Indeed, on a superficial view, we might naturally wonder that this tendency is not more conspicuously operative than is in fact the case. For even oneās mother tongue obviously must require to be learnt; and nobody learns his mother tongue so perfectly as never to make any grammatical mistake. In a language with a great variety of conjugations and declensions, mistakes of grammar mostly consist in assimilating the inflexion of the less common words to the more familiar types. We might therefore expect that, between forgetfulness and the instinct for consistency, the rarer conjugations and declensions would always rapidly drop out of use, and that all inflexional languages would in a few generations approach perceptibly nearer to the ideal state in which the same grammatical relation should always be denoted by the same change in the form of a word.
But in all matters of language the influence of tradition is extremely powerful. The mistakes or intentional innovations in grammar made by individuals are for the most part condemned by the community at large, and only few of them come to affect the general language. Probably most English children have sometimes said āmousesā or āspeaked,ā but these regularized forms do not appear in the speech of even illiterate adults. So the tendency to grammatical simplification in languages is usually slow in its working, unless it happens to be stimulated by some special cause.
Among the causes which hasten the progress of languages towards grammatical simplicity, there are two that require particular notice. There are (1) phonetic change; and (2) the mixture of peoples speaking different languages, or different dialects of the same language.
Phonetic Change.
When we study the history of any language, we always discover that, at some period or other, certain of its elementary soundsācertain āletters,ā as we might call them, of its spoken alphabetāhave undergone an alteration in pronunciation. The changes to which we here refer are unconscious and unintentional, and are so very gradual that it would need an acute and attentive ear to discern any difference between the sound of a word uttered by young men and by old men living at one time. But when, as is often the case, the pronunciation of a vowel or consonant becomes in each successive generation a little more unlike what it was at first, the total amount of change may in time be very great. If we could compare (by means of a phonograph or otherwise) the present pronunciation of some language with its pronunciation a few centuries ago, we might find, for instance, that all the aās had turned into oās, or all the dās into tās, or vice versa. More commonly, we should find that a particular vowel or consonant had changed into a certain other vowel or consonant whenever it occurred in the same part of a word (beginning, middle, or end); or whenever it came in an accented syllable; or whenever it came next to a certain other sound, or to any sound of a certain class; and that under other conditions it had either undergone a different kind of change, or else had remained unaltered.
The term āphonetic changeā is conventionally restricted to that kind of unconscious alteration of sounds which has just been described. If we study any particular language as it is spoken to-day, and ascertain what sound in it represents each of the sounds of some older form of the language under each of the varieties of condition under which it occurred, we shall obtain a body of rules which are called the phonetic laws of the present stage of the language. It is often said that the phonetic laws applicable to one and the same dialect and date have no exceptions whatever. Whether this is absolutely true or not, it is so nearly true that whenever we meet with a seeming exception we shall be pretty safe in believing that there has been at work some other process than āphonetic changeā in the sense above explained. For instance, it is not a case of phonetic change that we say āI broke,ā where our ancestors said āI brake.ā What has happened is not that a has changed into o, but that the old past tense has been superseded by a new one, imitated from the participle broken. Again, an apparent exception to a phonetic law may be due to the fact that one dialect has borrowed a form from another dialect in which the course of phonetic change had been different.
Why a particular phonetic change should take place in one language, dialect, or period and not in another is a question on which we cannot here enter. For our present purpose, it is enough to note the fact that the same original sound may develop quite differently in two dialects of the same language, and that a sound may continue for many centuries unaltered, and then enter on a course of rapid change.
The results of phonetic change, so far as they affect the history of grammar, are of three kinds:
1. Confluent development. Sometimes two originally different sounds come to be represented in a later stage of the language by a single sound.
Thus the Old English
and
(in certain positions) have yielded the Modern English
, so that
hl (whole) and
fla (foal) now form a perfect rime.
2. Divergent development. One and the same original sound may, owing to difference of conditions, yield two or more distinct sounds in the later language.
Thus in Old English
ic l de, I lead, and
ic l dde, I led, had the same vowel; but because in one word the vowel was followed by a single and in the other by a double
d, their modern forms have different vowels.
3. Dropping of sounds. In some cases the phonetic law relating to a particular vowel or consonant is that, when it occurs under certain conditions, it will neither remain unchanged nor change into anything else, but will vanish altogether.
Thus, an Old French t, if it comes at the end of a word, becomes silent in Modern French. Again, every short vowel which ended a word (of more than one syllable) in Old English has long ago dropped off, so that all the words which a thousand years ago were disyllables with short vowel endings are now monosyllables.
Now supposing that in any language the sounds which happen to be subject to these three kinds of phonetic change are those which are used in the inflexional endings, it is obvious that the result must be a considerable upsetting of the grammatical system. The effect, however, is not immediately to produce simplification. On the contrary, the tendency of ādivergent developmentā is to increase the number of declensions and conjugations, because the same original termination becomes different in different words. The effect of āconfluent developmentā and ādropping of soundsā is to make the inflexional system less efficient for its purpose by confounding different cases, persons, tenses, etc., under the same form. It is owing to changes of this sort in prehistoric times that the Latin language has the awkward defect of having only one form
(Musae) for the genitive and dative singular and the nominative and vocative plural of certain nouns. The same cause, also, accounts for the inconvenient peculiarity of Old English grammar, in having a large number of nouns with their nominative singular and nominative plural alike. This example is instructive, because it shows the fallacy of the notion sometimes maintained, that phonetic change does not destroy inflexions till they have already become useless. In what may be called prehistoric continental English, the plural ending of many neuter nouns was
. There came a time when it became a phonetic law that a final
always dropped off when it followed a heavy syllable, but remained after a light syllable. Hence in Old English as we know it the plural of
scip (ship) was
scipu, but the plural of
hs (house) was
hs, just like the singular. In this instance phonetic change produced two different effects: it made two declensions out of one, and it deprived a great many words of a useful inflexional distinction.
We thus see that the direct result of phonetic change on the grammar of a language is chiefly2 for evil: it makes it more complicated and less lucid. But when these inconveniences become too great to be endured, they provoke a reaction. The speakers of the language find out how to express needful grammatical distinctions by other than inflexional means; or else they generalize the use of those inflexional forms that have happened to escape decay, applying them to other words than those to which they originally belonged. In this way phonetic change leads in-directly to that kind of simplification which we shall find exemplified in the history of the English language.
Mixture of Peoples.
The second condition which we mentioned as favouring grammatical simplification was the mixture of peoples speaking different languages or dialects.
Let it be imagined that an island inhabited by people speaking a highly inflected language receives a large accession of foreigners to its population. To make the case as simple as possible, let it be further imagined that there is no subsequent communication with the outside world, and that nobody on the island can read or write. What may be expected to happen?
It is a matter of general experience that a person who tries to learn a foreign language entirely by conversation finds the vocabulary easier to acquire than the grammar. And it is wonderful how well, for the common purposes of intercourse, one can often get on in a foreign country by using the bare stems of words, without any grammar at all. Many Englishmen of the uneducated class have lived for years in Germany, and managed to make themselves fairly well understood, without ever troubling themselves with the terminations of adjectives or articles, or the different ways of forming the plural in nouns. In ...
