|Search Turner Comparative||Font Display||About the Dictionary||Front Matter||Copyright Information||Back to DDSA|
ALTHOUGH the form of this work was suggested by W. Meyer-Lübke's Romanisches Etymologisches Wörterbuch, the head-words under which the material is collected do not represent the vocabulary of a real and comparatively uniform language at all comparable with that of the actual Latin the history of which Meyer-Lübke followed into the modern Romance languages and which was in regular use for writing and administration over the whole Roman Empire and was the common speech of a large proportion of its inhabitants.
The phonetic systems of nearly all the New Indo-aryan languages are descended directly from that of the R̥gveda1 with one exception. Whereas Indo-europen k̂, ĝ, ĝh were represented by ś, j, h in the R̥gveda, in the Kafiri group they were retained in the intermediate stage of dental affricate as ts, dz; IE. gwh before palatal vowel which like ĝh was opened to h in the R̥gveda retained its occlusion as ǰ (ž) in Kafiri. It will therefore be realized that Kafiri words with these sounds are not to be derived immediately from the head-words under which they are shown: see, e.g., śíras-, dáśa, jānāˊti, *hadas-, hánati. One other possible divergence going back to a pre-Indo-aryan stage is the emergence in Middle Indo-aryan of (j)jh ~ kṣ as different developments of earlier *gzh; but both forms, having had wide expansion and not disclosing any dialectic boundaries in New Indo-aryan, will be found among the head-words with suitable cross-references: see, e.g., kṣárati and *jharati.
In some cases an older form has been replaced by an analogical creation, e.g. dūḍhīˊ- RV. (<*duẓ-dhī-) ~ durdhī- MBh. In other cases the original survives only in New Indo-aryan, e.g. *nīḍāti (<*niẓ-dā-) ~ nir-dātr̥- Mn., *būḍhi- (<*budzdhi-) ~ buddhi- Mn. In compounds the unexploded final -t of ut may be expected to be lost before following st(h), giving rise to uṣṭ(h)- attested in Middle and New Indo-aryan, although dialectically or through analogical replacement appearing only as utt(h)- in Sanskrit. Similarly ut before initial ś- appears most frequently in MIA. and NIA. as uśś- (uss-), whereas Sanskrit always has ucch-. To allow for the divergent development in such cases the head-word has been shown with *ut-sth-, *ut-ś-: see, e.g., *ut-sthāti, *ut-śīrsa-.
Generally speaking, however, the head-words have the phonetic structure proper to the earliest recorded form of the language which, while still comparatively uniform, was brought into India by the Aryan invaders. But even a cursory examination of Professor M. Mayrhofer's Kurzgefasstes etymologisches Wörterbuch des Altindischen will show that during the long history of this language as a literary medium in the form of Sanskrit a very large proportion of its vocabulary arose from internal development or came from external sources at a period when the various spoken languages derived from the Vedic speech had already greatly altered its phonetic form and grammatical structure. Many of the head-words, like so much of classical Sanskrit vocabulary, are in reality Middle Indo-aryan clothed, for the convenience of presentation, in an earlier phonetic dress. An example of this is provided by compounds with verbal prefixes: Oṛiyā uā `boiled sun-dried rice', Sinhalese avuva `sunshine' are, like Prakrit āyava-, derived from ātapá-, or Hindī āyat m. `sunshine' from ātapta-; but Khowar ātΛpik `to have high fever' must rest either upon a late MIA. *ātapp- (newly formed compound with ā from tappaï) or upon MIA. *āttapp- with analogical -tt- (after type ā-tt- < ā-tr-, etc.). The head-word
1The change of -ḍ- and -ḍh- to -ḷ- and -ḷh- was not universal.
ātapyatē under which the Khowar word appears is thus in reality a Middle Indo-aryan word in Old Indo-aryan form.
Owing to differences of dialect or of time at which many of the loanwords from Muṇḍā, Dravidian, or other languages came into Indo-aryan, they appear in forms which cannot be referred to a single Indo-aryan form, especially in the case of names of animals and plants. In such cases the New Indo-aryan words may be collected under more than one head-word with appropriate cross-references: see, e.g., mayūˊra-, śvāvídh-, kr̥muká-.
A particular class of words are those I have termed `defective'. These are adjectives which express, from language to language, almost any defect, whether physical, mental, or moral. They display, within a single language or from language to language, an almost bewildering variety of form, involving interchange of single with double consonant, aspirates with non-aspirates, voiced with unvoiced, dental with retroflex, nasalized vowel with unnasalized, vocalization with a or i/ē varying with u/ō: see *bukka-4 and the list of similar series there given. The variety of form is no doubt due to a variety of causes: they are liable as `expressive' words to changes which indicate emphasis, as derogatory words to deliberate deformation, and under both headings to the formation of rhyme-words. While some, e.g. baṇḍá-, are found in Vedic, and some, e.g. kuṇṭha-, have parallels in Iranian, and some, e.g. *runda-, possibly derive from Indo-european, a large number appear to come from non-Aryan sources—Dravidian and especially Muṇḍā—and may owe some of their diversity of form to different dialects within those language-groups (see in the last instance F. B. J. Kuiper in Lingua 14, 54-86). There are parallel series of words meaning `lump' with the same variation of form: see, e.g., *lakka-2 ~ *lakka-1.
In the spelling of new Indo-aryan words, the practice of the particular dictionaries and vocabularies from which they have been taken has in general been followed. This has led to a number of inconsistencies: for example, the confusion of long and short e and o, or the unequal differentiation between palatals and dental affricates (c, j ~ ċ, j̈) in the Pahāṛī languages. In some cases where a dictionary confuses phonetic with traditional spelling, the phonetic has been used throughout in transcribing from devanāgarī: thus in Assamese the devanāgarī palatals, both non-aspirate and aspirate, have been regularly transcribed as s and z. For the Kafiri and Dardic languages (in which Morgenstierne, like Lorimer, has used č and ǰ for palatals, c and j for dental affricates) I have retained č and ǰ (as also for Gypsy), and have used ċ and j̈ for dental affricates, c̣ and J̣ for retroflex. The acute accent Q, which is used in Sanskrit words for a tone (udātta), indicates a stress when used in NIA. words; the grave Q , which is the svarita of Sanskrit, denotes a falling tone in NIA. words.
The head-words, numbered for subsequent reference in the indexes, are printed in heavy type, for example 1100 āˊñjana-. Its descendants in Middle and New Indo-aryan are printed in italic: Pa. añjana-, Ash. aná, etc. Any form within an article printed in small capitals, e.g. AJÑĀNIN-, will be found as a head-word in its alphabetic place. So too the words in small capitals at the end of an article which record the derivatives of the head-word (e.g. ĀÑJANĨYA-) or the compounds in which it occurs (e.g. ŚŌBHĀÑJANA-). Discussion of the earlier history of the head-words, especially where the more modern forms throw light upon it, or of their connexion with other head-words is contained in the square brackets. If, as in this instance, these contain a root (e.g. √AÑJ), this will be found in its alphabetic position in heavy capitals (AÑJ) followed by all forms which can be grouped as belonging to the root and which have been or will be entered as head-words in their due alphabetic places (e.g. AKTÁ-, AÑJÁTI, ĀˊÑJANA-, ÚPĀKTA, VYÀKTA-, *SAMAÑJATI, etc., etc.).
The order in which words from new Indo-aryan languages are quoted is purely geographical, being based on their present distribution rather than on any special dialectic connexions among them. Beginning in the West with European dialects of Gypsy, it moves through the Ḍumāki of Hunza (like Gypsy, belonging originally to a language group of India proper) to Kafiri and Dardic, Kashmiri, Sindhī, Lahndā, Panjābī, along the Himalaya to West and Central Pahāṛī groups, to Nepāli and Assamese, south to Bengali and Oṛiyā, westward again across the Gangetic plain through Bihārī, Maithilī, Bhojpurī, Hindī to the Rājasthānī dialects, south again through Gujarātī to Marāṭhī and Koṅkaṇī, and finally to Sinhalese (whose closer linguistic connexion is rather with the eastern languages, especially Oṛiyā) and out in the Indian Ocean to the dialect spoken in the Maldive Islands.