Spanish.
Quien canta sus males espanta.
quem cantat suos malos expav(ere).
i.e. He who sings frightens away his ills.
Por la calle de despues se va á la casa de nunca.
per illam callem de de-ex-post se vadit ad illam casam de nunquam.
i.e. By the street of by and by one goes to the house of never.
French.
Un tiens vaut mieux que deux tu l’auras.
unum tene valet melius quod duos tu illum habere-habes.
i.e. One take-it is worth more than two thou-shalt-have-its.
Parler de la corde dans la maison d’un pendu.
parabola de illam chordam de intus illam mansionem de unum pend(o).
i.e. (Never to) talk of a rope in the house of a hanged man.
It is plain on the face of such sentences as these, that Italian, Spanish, and French are in fact transformed Latin, their words having been gradually altered as they descended, generation after generation, from the parent tongue. Now even if Latin were lost, philologists would still be able, by comparing the set of Romance languages, to infer that such a language must have existed to give rise to them all, though no doubt such a reconstruction of Latin would give but a meagre notion, either of its stock of words or its grammatical inflexions. This kind of argument by which a lost parent-language is discovered from the likeness among its descendants, may be well seen in another set of European tongues. Let us suppose ourselves listening to a group of Dutch sailors; at first their talk may seem unintelligible, but after a while a sharp ear will catch the sound of well known words, and perhaps at last whole sentences like these:—Kom hier! Wat zegt gij? Hoe is het weder? Het is een hevige storm, ik ben zeer koud. Is de maan op? Ik weet niet. The spelling of these words, different from our mode, disguises their resemblance, but as spoken they come very near corresponding sentences in English, somewhat old-fashioned or provincial, thus:—Come here! What say ye? How is the weather? It is a heavy storm, I be sore cold. Is the moon up? I wit not. Now it stands to reason that no two languages could have come to be so like, unless both were descended from one parent tongue. The argument is really much like that as to the origin of the people themselves. As we say, these Dutch and English are beings so nearly alike that they must have descended from a common stock, so we say, these languages are so like that they must have been derived from a common language. Dutch and English are accordingly said to be closely related to one another, and the language of Friesland proves on examination to be another near relative. Thence it is inferred that a parent language or group of dialects, which may be called the original Low-Dutch, or Low-German, must once have been spoken, though it is not actually to be found, not happening to have been written down and so preserved.
Now it is easy to see that as ages go on, and the languages of a family each take their separate course of change, it must become less and less possible to show their relationship by comparing whole sentences. Philologists have to depend on less perfect resemblances, but such are sufficient when not only words from the dictionary correspond in the two languages, but also these are worked up into actual speech by corresponding forms of grammar. Thus when Sanskrit, the ancient language of the Brahmans in India, is compared with Greek and Latin, it appears that the Sanskrit verb dâ expresses the idea to give, and makes its present tense by reduplicating and adding a person-affix, so becoming dadâmi, nearly as Greek makes didōmi: from the same root Sanskrit makes a future participle dâsyamânas, corresponding to Greek dōsomenos, while Sanskrit dâtâr matches Greek dotēr = giver. So where Latin has vox, vocis, vocem, voces, vocum, vocibus, Sanskrit has vâk, vâćas, vâćam, vâćas, vâćâm, vâgbhyas. When such thoroughgoing analogy as this is found to run through several languages, as Sanskrit, Greek, and Latin, no other explanation is possible but that an ancient parent language gave rise to them all, they having only varied off from it in different directions. In this way it is shown that not only are these particular languages related by descent, but that groups of ancient and modern languages in Asia and Europe, the Indian group, the Persian group, the Hellenic or Greek group, the Italic or Latin group, the Slavonic group to which Russian belongs, the Teutonic group which English is a member of, the Keltic group which Welsh is a member of, are all descendants of one common ancestral language, which is now theoretically called the Aryan, though practically its nature can only be made out in a vague way by comparing its descendant languages. Some of these have come down to us in forms which are extremely ancient, as antiquity goes in our limited chronology. The sacred books of India and Persia have preserved the Sanskrit and Zend languages, which by their structure show to the eye of the philologist an antiquity beyond that of the earliest Greek and Latin inscriptions and the old Persian cuneiform rock-writing of Darius. But the Aryan languages even in their oldest known states had already become so different that it was the greatest feat of modern philology to demonstrate that they had a common origin at all. The faint likeness by which Welsh still shows its relationship to Greek and German may give some idea of the time that may have elapsed since all three were developed off from the original Aryan tongue, which itself probably ceased to exist long before the historical period began.
Among the languages of ancient nations, another great group holds a high place in the world’s history. This is the Semitic family which includes the Hebrew and Phœnician, and the Assyrian deciphered from the wedge-characters of Nineveh. Arabic, the language of the Koran, is the great modern representative of the family, and the closeness with which it matches Hebrew may be shown in familiar phrases. The Arab still salutes the stranger with salâm alaikum, “peace upon you,” nearly as the ancient Hebrew would have said shâlôm lâchem, that is, “peace to you,” and the often-heard Arabic exclamation bismillah may be turned into Hebrew, as be-shêm hâ-Elohim, “in the name of God.” So the Hebrew names of persons mentioned in the Bible give the interpretation of many Arabic proper names, as where Ebed-melech, “servant of the king,” who took Jeremiah out of the dungeon, bore a name nearly like that of the khalif Abd-el-Melik, in Mohammedan history. But no one of these Semitic languages has any claim to be the original of the family, standing to the others as Latin does to Italian and French. All of them, Assyrian, Phœnician, Hebrew, Arabic, are sister-languages, pointing back to an earlier parent language which has long disappeared. The ancient Egyptian language of the hieroglyphics cannot be classed as a member of the Semitic family, though it shows points of resemblance which may indicate some remote connexion. There are also known to have existed before 2000 B.C. two important languages not belonging to either the Aryan or Semitic family; these were the ancient Babylonian and the ancient Chinese. As for the languages of more outlying regions of the world, such as America, when they come into view they are found likewise to consist of many separate groups or families.
This slight glimpse of the earliest known state of language in the world is enough to teach the interesting lesson that the main work of language-making was done in the ages before history. Going back as far as philology can take us, we find already existing a number of language-groups, differing in words and structure, and if they ever had any relationship with one another no longer showing it by signs clear enough for our skill to make out. Of an original primitive language of mankind, the most patient research has found no traces. The oldest types of language we can reach by working back from known languages show no signs of being primitive tongues of mankind. Indeed, it may be positively asserted that they are not such, but that ages of growth and decay have mostly obliterated the traces how each particular sound came to express its particular sense. Man, since the historical period, has done little in the way of absolute new creation of language, for the good reason that his wants were already supplied by the words he learnt from his fathers, and all he had to do when a new idea came to him was to work up old words into some new shape. Thus the study of languages gives much the same view of man’s antiquity as has been already gained from the study of races. The philologist, asked how long he thinks mankind to have existed, answers that it must have been long enough for human speech to have grown from its earliest beginnings into elaborate languages, and for these in their turn to have developed into families spread far and wide over the world. This immense work had been already accomplished in ages before the earliest inscriptions of Egypt, Babylon, Assyria, Phœnicia, Persia, Greece, for these show the great families of human speech already in full existence.