Some Sanskrit and Arabic Words and the Puzzles They Create [Archives:2000/13/Culture]
March 27 2000
Dept. of English
Taiz university
Words, history informs us, tend to travel, often far and away from their birthplace meandering through continents and centuries in different shapes and senses, unrecognizable to untrained eyes and incurious minds. Outside nature, such words are the most interesting chameleons, which challenge and sometimes defy, successfully the scholarly ingenuity of the word-trackers whose mission in life is tracing and reconstructing their original shape in addition to deciphering their primordial sense. However, in most cases, the methods and techniques developed by the etymologists (that is, professional word-trackers) have, time and again, proved effective in reaching out far and deep into the history of such wandering words in order to identify their provenance and etymons. Nevertheless, conundrums and enigmas are not infrequently met with in this field of study as in others demanding greater rigor of research and deeper understanding of the nature of problems. Indeed, the existence of such problems in any discipline stimulates the refinement of its techniques of study and if old techniques fail, search for new methods of study is keenly pursued leading to fresh and more insightful researches. And of course their solution necessarily results in the extension of the horizon of the discipline itself. This is a report on a few riddles (so they seem to me at least) which involve some words in Arabic and Sanskrit and the answer to which might possibly entail revision of the received opinion on the relation between Arabic and Sanskrit.
By courtesy of Yemen Times, I have had the opportunity in the past of reporting to its readers my investigations into the travels of a number of Arabic words in times past and their naturalization in Hindi, one of the major languages of the Indian sub-continent. I have also reported on a number of words of Indian origin, especially Sanskrit/ Hindi, which have established themselves in Arabic and now form past of Arabic lexicon. To the best of my knowledge and belief, these reports based on the sound principles of etymological science are accurate and convincing. But in the course of my further investigations in the exchange of words between Arabic and Sanskrit/ Hindi I have encountered a few rather puzzling problems, which have been simply stated as here under. Let us begin by first considering the following pair:
(1) Sanskrit Kalama: Arabic ()
Sir Monier-Williams ‘A Sanskrit-English Dictionary’ describes the meaning of Sanskrit of Kalama as follows: (i) a sort of rice (sown in May and June and ripening in December or January); (ii) a reed for writing with. The word is attested to have been used in these senses in ancient Sanskrit texts referred to by sir Monier- Williams. In addition, he cites cognates from Greek and Latin, the two major languages of the Indo-European family. The Greek word Kalamos and the Latin word Calamus correspond to the Sanskrit word Kalamas; all three are nominative singular forms. That is Sanskrit, Greek and Latin agree on the form and meaning of the word in question; all three appear to have inherited it from the parent language, Proto-Indo-European. Naturally, the presumption is unavoidable that it is a very old Indo-European word dating back to the period when Latin, Greek and Sanskrit were one language; that is, they had not split from the parent language and separated from one another. In any case, the closeness of genetic relationship between the three language, well-attested on other grounds, is beyond any doubt. Whether it be a lexical item or a grammatical process which is shared by Sanskrit, Greek and Latin, it must be presumed to have been inherited from the common stock and cannot be reasonably considered in all three. At the same time it cannot be considered a case of later borrowing in any of the languages from the other; the odds are heavy and far too many against such an assumption.
To be sure, there is another word in Sanskrit Lekhani meaning “an instrument for writing or painting, read-pen painting-brush, pen, pencil”; obviously, it is similar in meaning to Sanskrit Kalama, but it is obviously a derived rather than primitive word. It is derived from the root Likh, and it is an innovation in Sanskrit in this sense.
Interestingly, the same dictionary of Sir Monier-Williams carries the Arabic word qalam by the side of Sanskrit kalama, but does not define it; that is, it does not state the meaning(s) of qalam. But the fact that it appears with Sanskrit kalama seems to suggest that the lexicographer was most certainly aware of the partial similarity between the form and meaning of Arabic galam and Sanskrit kalama. Similarly between the two words is partial inasmuch as Arabic galam {k}; {g}, in phonetic terms, is a uvular consonant while {k} is a velar consonant. In respect of meaning, too, the similarity is partial, for Arabic qalam (), to be sure, has the meaning “reed pen, pen”, but has other meanings besides. Even so, the partial similarity is so striking that it cannot but prompt us to seriously entertain the hypothesis of borrowing from Indo-European language into Arabic. Specifically, qalam () may have been borrowed into Arabic from Latin or Greek, if not from Sanskrit of course it is open to someone to propose that Indo-European languages borrowed from Arabic, but then he would have to prove that the three Indo-European languages borrowed from Arabic at the same time with the same modification, phonetic and semantic, only one word, and this process has not recurred since. In either case assuming it to be a case of borrowing, the borrowing must date back to a very distant past, to prehistoric times, perhaps. But if we are prepared to travel to prehistoric times, we can also postulate that Arabic qalam () like Sanskrit kalama, Latin calamus, Greek kalamos continues the Proto-Indo-European word, but that can be done only on the hypothesis that Semitic and Indo-European languages had at one time a common ancestor a hypothesis that files in the face of received opinion in the field of historical and comparative linguistics of different language families. Thus, borrowing remains the only plausible hypothesis to account for the presence of Sanskrit kalam and Arabic qalam (); but how do we decide the direction of the borrowing?
Our next pair of words is:
(2) Sanskrit vadana: Arabic badan ().
Sanskrit vadana, according to dictionaries, means “the face, mouth, countenance”. In this sense the word has occurred in a number of ancient Sanskrit texts ranging from literacy to scientific.
Arabic badan, again according to dictionaries, has the sense of ‘body, trunk, torso’. It too, is an old word attested in old Arabic texts.
The two words show partial similarity in form meaning. In form they differ only in respect of the first consonant: Sanskrit /v/ as against Arabic /b/; Sanskrit Labio-dental semi-vowel as against Arabic bilabial. For the rest they would be generally considered identical. In meaning they differ as a part of the body, Arabic badan. They are similar inasmuch as they name a part of or the whole of the human body; they belong to the same semantic field, and quite closely connected.
Now, at some time in the history of Sanskrit a phonetic change seems to have set in according to which /v/ in word-initial position was replaced by /b/. As one Sanskrit dictionary puts it: “it (i.e. v) is often confound and interchanged with labial /b/”. For example, vakeruka into bakeruka, and so did many without any accompanying change in the meaning of these words. It was purely a phonetic change’ the lexical items affected by the change were in free variation. By the time Sanskrit gave way to Hindi, one of the daughter languages of Sanskrit, the change became much more extensive and /v/ in word-initial position of a large number of Sanskrit words came to be systematically replaced by /b/.
The point of this digression is to show how Sanskrit vadana was replaced by Hindi badana. However, at this stage, its meaning remained unchanged. That is, it continued to mean ‘the face, mouth, countenance’; the old form changed, but its old meaning remained what it was earlier.
At this point in the history of Hindi, a number of Arabic words were borrowed into Hindi, and it is reasonable to believe that one of these words was badan meaning ‘body, trunk, torso’. Thus, a pair of homophones came into being, which had two different sources and involved two different linguistic processes. However, in modern Hindi, at any rate, the two have merged together and have come to mean: ‘body, trunk, torso’. But the question is: Is the sound shape of badan a borrowing from Arabic, or is it the meaning alone which has been borrowed? Perhaps, it makes no difference to those who use the word today, but to him who is interested in the travels of words across time and space, it does matter. However, it is rather hard for him to decide the direction the borrowing has gone without some more evidence than is available at the moment.
The third pair of Sanskrit and Arabic words that we propose to consider next is:
(3) Sanskrit ma: Arabic ma ().
This pair differs from the preceding ones in being a pair of structural words in contrast to the pairs of content words discussed above. The content words are defined in a dictionary in terms of their referential meanings; structural words, on the other hand, are described in terms of their grammatical functions in the language. For example, prepositions and conjunctions of English are structural words in the grammar of English in contrast to nouns and verbs which refer to objects, states and actions in the world. It is a widely held view based on observation and experience that content words are more reality borrowed or lent than structural words, which are firmly integrated into the grammatical structure of the language.
The purpose of these general observations it to provide a perspective on the nature of the problem we are two different, genetically unrelated, geographically of ma in Arabic and in Sanskrit, not only closely similar in form but also similar in grammatical function demands explanation. A plausible explanation may be suggested by assuming that one of the two languages has borrowed from the other; but ma being a structural word, borrowing by either of the two languages makes it rather unlikely. So how do we account of ma in Sanskrit and ma in Arabic? Let us first note the details of similarities.
Sanskrit ma is an indeclinable, invariable negative particle; it is a particle of negation and prohibition. It is cognate with Greek negative particle me. Sanskrit ma commonly occurs in the earliest stages of Sanskrit with the past tense forms (i.e. ÉÉforms in particular) of verbs in subjunctive mood. At the later stages of Sanskrit, however, when Vedic Subjunctive seems to have fallen out of use, ma is more frequently constructed with the imperative forms of verbs. Sometimes it is also used in place of the simple negative na used to deny a statement.
Arabic ma is also an indeclinable, invariable negative particle; it usually occurs with the perfect forms of Arabic verbs, which are comparable to the artist forms of Sanskrit verbs. For example, ma kataba () ‘He did not write’ ma dahaba () ‘ He did not go’, and so on. The negativising function of ma is quite clear in constructions in which it occurs with the perfect forms of verbs like zala () ‘To cease’, ada () ‘To return’ or bagiya () ‘To remain’. Some examples are: ma zala Hasuun dahiban () ‘Hasan did not cease going (i.e. Hasan is still going); ma ada yarjau () ‘He did not return again; ma bagya ahadun () ‘No one remained.
However, Arabic ma differs from Sanskrit ma insofar as it does not occur in construction with imperative forms of verbs. Even so the similarities between Sanskrit ma and Arabic ma are quite impressive. They need to be accounted for. If we seek to account for them by assuming separate and simultaneous developments towards similar ends in two genetically unrelated languages independent of each other, it would amount to making teleology a very important regulative principle operative in language systems. It would also implicitly deny the principle of arbitrariness widely held to characterize the relationship between linguistic forms and their and their meanings and functions. Thus, we are led to assume borrowing as the most plausible mechanism accounting for the similarities between Arabic ma and Sanskrit ma. But the question is: Which way did the borrowing g? The present writer does not claim to know it as yet.
The next pair in this series consists of:
(4) Sanskrit gard: Arabic garad ().
Let us begin by nothing the formal similarities and differences between the two words. Phonetically speaking, Sanskrit gard begins with a voiced velar stop consonant, while Arabic garad begins with a voiced velar/ uvular fricative consonants. The differences in respect of the vowels in them are not substantive; for the most part they are a consequences of different lexicographical practices with regard to citation forms. For example, Sanskrit dictionaries cite the root form of verb, while Arabic dictionary cite the third person singular number perfect form of verb. Consequently, Sanskrit gard is the root form whereas Arabic garad is the third person singular number of the perfect form of the verb.
As regards the meaning of the Sanskrit garad, one dictionary states the following: ‘To shout, give shout of joy, to emit any sound’. The meaning of Arabic garad according to one dictionary, is: ‘to sing, twitter (bird), warble’. Obviously, the meanings are not identical; identity of meanings from two languages is a myth; nevertheless, there is a large measures of overlap between the meanings of Sanskrit gard and Arabic garad. Now, the similarities between the forms and meanings of the words under consideration here cannot be plausibly attributed to chance coincidence. At the same time these two words cannot be considered to be continuations, with certain modifications, of an inherited word from an earlier pre-historic common parent language. Sanskrit and Arabic are not deemed to have descended from a common parent language. Thus, we are left with the third possibility of borrowing from one language to another, but again, as before there is no way we can as certain direction of borrowing: did Arabic borrow garad from Sanskrit gard, or Sanskrit from Arabic, with appropriate modifications in sounds and meanings?
The preceding discussion of a few pairs of Sanskrit-Arabic words has tried to account for their similarities in form, function and meaning, but has, in each case, been constrained to reach the conclusion that no assumption other than that of borrowing appears plausible enough to be considered for the purpose of accounting for the similarities noted above. Even on this assumption, however, it has not been possible to decide the direction of the borrowing between Sanskrit and Arabic. Which one has borrowed from which other?
In contrast to the preceding Sanskrit-Arabic words of undecidable provenance, we now turn to those Sanskrit-Arabic words those similarity in form and meaning can be effectively decided in terms of borrowing. We shall begin with:
(5) Sanskrit vidaya: Arabic wadaa ()
One dictionary of Sanskrit defines the meaning of vidaya as follows: (1) “division, distribution; (2) permission to go away, dismissal with good wishes ( in this sense perhaps not a Sanskrit word”. The dictionary goes on to cite the Arabic word wada (). Now Arabic wada () carries the following meaning: “farewell, leave-taking, good-bye”. It is obvious that the second meaning given in Monier-Williams dictionary of Sanskrit for the entry vidaya is very similar to the meaning of wada () given in Arabic dictionaries. However, the form of the Sanskrit word vidaya, and the verbal root from which it is derived are authentically Sanskrit. It seems reasonable to conclude that the close similarity between Sanskrit vidaya and the spoken form of Arabic wada () might have triggered the transfer of the Arabic meaning to the Sanskrit word-form. Alternatively, it might be argued that only the meaning but the form also of the Arabic word wada () could have been borrowed into Sanskrit itself, with certain sound modifications, of course. That there has been an awareness of the alieness of the meaning is evident from the attempts made by certain Sanskrit to replace this word with this meaning by neologisms, but without much success in popular usage. In this case then we can be quite certain that the meaning of Sanskrit vidaya is a borrowing from Arabic, if not the form of it.
We shall consider now another pair of:
(6) Sanskrit lajavarta: Arabic lazaward ()
lazaward () is an Arabic word. Arabic dictionaries carry this word along with some of its derivation, like (i) lazawardi () and (ii) lazawardiya (). We shall be concerned here with lazaward (), this form and meaning, simply because (i) and (ii) are transparently derived from it by means of suffixes incontestably Arabic in origin. In this sense lazaward () is basis in form as well as in meaning.
The interestingly enough, a Sanskrit lajavarta too, which alternates with rajavarta. This alternation is sanctioned by the grammar of Sanskrit and it characterizes a sizable number of Sanskrit words. These two variant forms have one and the same signification or meaning. The meaning of lajavarta or rajavarta found in a dictionary of Sanskrit is: ” lapis lazuli”, exactly the same meaning as of Arabic lazaward. The Sanskrit dictionary describes the meaning of this word at some laugh; it says: “lajavarta is a kind of diamond or other gem (of an inferior quality said to come from the country of virata, and regarded as lucky possession though not esteemed as an ornament”.
The similarity between Arabic and Sanskrit words, despite the differences between some of their constituent sounds, is striking, both in form and meaning. We cannot attribute such a similarity to accident. Given the principle of the arbitrariness of linguistic signs, accidental similarity between words of the size of lazaward () seems rather highly improbable. But, at the same time, it cannot be due to their common origin either. Thus, borrowing remains the only mechanism which might have brought about the similarity between Arabic lazward () and Sanskrit lajavarta. But before we decide on the direction of the borrowing, let us note some of the unusual features of Arabic lazaward (. First, it does not follow the canonical syllabic pattern of words in Arabic; next, Arabic dictionaries do not carry the verb root from which this noun can be derived. On these two counts of its phonological structure and grammatical construction lazaward appears quite alien to the Arabic language. It is an isolated word in Arabic.
On the other hand, lajavarta conforms to the syllabic and grammatical structuring quite common to a large number of Sanskrit words. Therefore, we can reasonably conclude that lajavarta is original with the Sanskrit language, and that it has been borrowed into Arabic with some sound modifications. In this case then the fact of borrowing and the direction of borrowing do not seem liable to controversy.
The pair we shall next consists of:
(7) Sanskrit krimija: Arabic qirmiz ()
obviously, a deal of similarity exists between the surface forms of the two words mentioned in this section. Sanskrit krimija has a number of meanings, but the one that concerns us here is: “Lac red dye caused/ produced by insects”. Very similar to it is the meaning of Arabic qirmiz (): that is, “the dried bodies of the female kermes insect, coccus ilicis, which yield a red dye stuff”. Qirmizi () is of course derived from qirmiz () its meaning in English is ‘crimson, carmine, scarlet’. In fact, English “crimson” is ultimately traced to Arabic qirmizi () in etymological dictionaries. Besides, the cognates of Sanskrit krimija have been identified in a number of Indo-European language, like Lithuanian, Russia, Latin, Gothic and Irish.
We however, intend to decide here whether Sanskrit krimija and Arabic qirmiz () owe their close similarity in form and meaning to chance coincidence, common origin or to borrowing. Chance coincidence is inconceivable as an explanation of the phenomenon under consideration; it simply short-circuits the inquiry. Common origin is not demonstrable yet. Consequently, we are led to consider borrowing as the only reliable process which could account for the observed similarity between Sanskrit krimija and Arabic qirmiz ().
As in the case of lazwarad (), we argue that Arabic qirmiz () is an isolated word in Arabic. It is abnormal in respect of its syllabic structure; it occurs as a noun in Arabic, but there is no verb root from which we can derive it. On the other hand, Sanskrit krimija is one conforms to the regular phonological, grammatical and semantic ordering of elements and parts characteristic of Sanskrit. Indeed, Sanskrit krimija can be easily analyzed into two parts: krimi and ja. The two parts have their own separate and identifiable meaning: krimi means ‘insect’ and ja means ‘produced from by’. The meaning of the whole word can be satisfactory computed from the meaning of its parts. In other words, the meaning of krimija in Sanskrit is clearly motivated, it is not arbitrary, which is the case with Arabic qirmizi (). Therefore, we are inclined to conclude that Arabic qirmiz () may be Sanskrit krimija borrowed into Arabic from Sanskrit.
——
[archive-e:13-v:2000-y:2000-d:2000-03-27-p:./2000/iss13/culture.htm]
