The Script of Harappa and Mohenjodaro and its Connection with Other Scripts c10.s28: Analysis of Tables XXXI - XXXVII
The Script of Harappa and Mohenjodaro and its Connection with Other Scripts c10.s28
- Language
- English
Analysis of Tables XXXI - XXXVII
c10.s28.1
In this table we have the numerical signs from 3 to 9. The principal evidence that these are numerical signs is their remarkable correspondence with the same numerical signs in Proto-Elamite and Sumerian. It will be noted that the commonest occurrence of these signs is with the tree sign. All of them without exception are found with this sign (in the case of 8 the numerical is compounded with Y doubtless for phonetic reasons). With some this combination forms the majority of the total number of occurrences of the numeral sign. It has been already suggested that we may regard this combination as the sign plus the ordinal suffix. Indeed in view of the fact that this combination is found as a complete text, presumably a proper name, in Nos. 2, 17, 18, 25, 72, 76, 83, 113, it is difficult to conceive any other explanation. Doubtless " and ' were originally numerical signs, but they do not appear in any contexts that will bear a numerical interpretation in our texts. They were used to represent the vowels Y, I, with which their numerical values were perhaps homophonous, and to avoid confusion the place of " as a numeral seems to have been taken by || , at least when the numeral sign for '2' was required in a proper name. This at least is what appears to be suggested by the evidence of Table XXXVI, Nos. 84-87, especially 84, where we get Y|| appearing as a complete text just like Nos. 2, 17, 18, etc. mentioned above. It will be observed that sometimes this || resumed its original size " , and was then written in the middle of the line (see Table XXXVI, Nos. 123-125). But this would not be easy to distinguish from the vowel " which was also occasionally written towards the middle of the line (See Table XXX, Nos. 1-7), and consequently the elongated form || appears to have been normally adhered to. The form | is probably the numeral equivalent of ' , but does not appear to be used in a numerical sense in these texts except in No. 45 of Table XXXVII. To decide whether in a given text a numeral sign is to be read as a numeral or as a word or syllable that happens to be a homophone of that numeral we have two indicators: (a) the recurrence of a particular sign accompanied by several different numerical signs, (b) the recurrence of one numeral sign, and one only, a number of times with one and the same non-numeral sign. In the former case the numeral sign is to be read as a numeral, in the latter as a homophone unconnected with any numeral except by the accident of phonetic identity. There will remain a number of cases where a given sign is found only once or twice with a numeral sign. These will remain for the present dubious.
c10.s28.2
Applying the above criteria we find that when a numeral sign is followed by Y, ☥, ☥, ) X, ⊐ it is to be read as a numeral, where the numeral Ⅲ is followed by ☥; where the numeral Ⅲ follows ⊐; where it precedes ⊐; where the numeral ‖ precedes ⊐, ⊐, ↑ or follows ⊐ in the sequence ⊐ ‖ ⊐; in all cases where | occurs except M. 133; it is not to be read as a numeral.
c10.s28.3
Table XXXVI, No. 88 is very interesting; it shows that ‖ written in the middle of the line can be substituted for ‖ and is thereby sharply differentiated from ‖ the vowel. The vowel ‖ and the sign ‖ are distinct in the script, and it is perhaps scribal carelessness, or perhaps a survival of the era before their differentiation, that accounts for both occasionally appearing as ‖. They are however never found both written this way on one and the same text.
c10.s28.4
With regard to Ⅲ, Ⅲ, Ⅲ an examination of the plates will show that the length of the strokes varies considerably, especially with the "tree" sign. With ⊐ the length of the accompanying Ⅲ is apparently constant. The difference in length is probably determined merely by convenience and an aesthetic consideration for the appearance of the line. So that we may almost certainly regard Ⅲ and Ⅲ, Ⅲ and Ⅲ, Ⅲ and Ⅲ as mere graphic variants.
c10.s28.5
The signs in Tables XXXIV and XXXV are probably not of numerical origin. Since (a) it is unlikely on Sumerian or Proto-Elamite analogy that the digits were represented by single strokes exceeding 9 in number, (b) there are no intervening signs of 10 and 11 strokes, (c) the signs are not found in the normal numerical sequences, notably Y R. The pictographic origin of these signs is perhaps to be found in the ornamental design on the symbol [Figure:] (see Plate I). If this be so it is quite possible that the three signs in Col. IV of Tables XXXIV and XXXV should be regarded as simple variants.
c10.s28.6
It is possible that in some cases at least |||| , ||||| (Table XXXVI) |||||| , ||||||| are to be regarded not as numeral signs but as simple variants of a single sign of non-numerical origin whose pictographic original has been lost. Perhaps a fence? This is suggested by a comparison of texts 82, 92, 93 of Table XXXI, and of texts 2-5 of Table XXXVI. The normal way of writing the numerals would appear to be | ; || ; ||| ; |||| ; ||||| (rarely ||| ) ; |||| ; ||||| ; |||||| ; ||||||| . It is not certain how 10 was written.
c10.s28.7
That which is found on the Harappa prisms is the numeral 10, and that the prisms themselves (which usually contain names identifiable with those on the seals on one side) are receipts of tribute, etc. |' , Table XXXII is not likely to be a graphic variant of ||| numeral, since in the uneven numbers the majority of strokes is always placed in the upper layer as in Sumerian and Proto-Elamite.
c10.s28.8
' ' , Table XXXIII, is not likely to be a graphic variant of ' ' since it appears in no normal numerical sequence. It may be compared in shape with the Proto-Elamite ' ' (See D.E.P. XVII. Tablets passim, usually in the first column, but never initial as here in texts 5, 6).¹ We may now consider ' ' in texts 144-116 of Table XXXI. Inasmuch as ' ' is found preceding ' ' in No. 112, where it is clearly phonetic (' ' is preceded by no other numeral of the Table), (2) ' ' is only found in 5 instances, Nos. 114-116, 118, 120, in all of which it is separated from other combinations of short strokes by only one intercalated sign, (3) in 116, which seems to be an identical word with those contained in 114 and 115, the order of the strokes is reversed without apparent derangement of the sense, I conclude that ' ' and ' ' are but ' ' combined. Similarly ' ' is but ' ' combined (doubtless on phonetic grounds). It is significant that ' ' and ' ' are normally preceded by numeral signs. Similarly ' ' is ' ' , ' ' is ' ' ; which again is what we should expect since elsewhere ' ' is preceded by numeral signs. And ' ' is ' ' . With regard to the sign ' ' in Table XXXI. It might be taken for a divided form of ' ' if any such sign existed. But as we have just shown it does not. '4' is always written ' ' . There is no sign of which ' ' could be a form modified for purpose of combination. Moreover it is quite certain that in the case of ' ' at least we have no compound of the ordinary sort. For from all we have seen of compounds the enclosed portion is always to be read last. Now in the case of Nos. 45 and 46 this would break up two well established sequences. Here the element ' ' can only represent a modification of or addition to the final syllable of the words ' ' , ' ' . In these two cases the modifying element 𐀁 would appear to make no difference to the sense, but to be merely euphonic due perhaps to these words being final: they are final nowhere else (see Table XIII). In the case of No. 41, 𐀁 seems to affect the sense as well as the sound of 𐀂, which would appear to be a name in itself. 'To 𐀃; The son of 𐀄'. 𐀂 is nowhere else found final. The same remarks apply to the occurrence of 𐀁 with 𐀂. A comparison of Table XIII, 183 and 181 with 195 and 197 would indicate that in these cases 𐀂 and 𐀅 were identical in sense, and if not identical at least interchangeable in sound. It will be observed that in the fragmentary text 53 𐀅 is final: elsewhere 𐀂 is never final - or solus. Here then there would appear to be a change of sense. In text 55 𐀆 is also final. It is only once final elsewhere (Table XIII, 297). In text 58 𐀇 is final, and solus; in 61 it is also final. Elsewhere 𐀇 is only once final (Table XXI, 34). It is noteworthy not one of the sequences found with 𐀇 are found with 𐀈, notably 𐀉 𐀇 : so perhaps 𐀈 is modified in sense as well as sound. 𐀊 is final in every occurrence; 𐀋 is never final. is final in No. 66. 𐀍 is never final.
c10.s28.9
¹ Cf. also Brahmi ' ' = ' ' . It is possible that this may be an alternative writing for ' ' when i is used as a full vowel syl-
c10.s28.10
What sort of a graphic modification then is this which usually alters the sense, but not always; which alters the sound but slightly; which has a marked tendency to attach itself to certain syllables when final? An indication may be obtained through Table XXXIII, No. 9. Here we have a sign 𐀎 which is almost certainly to be identified with 𐀏, seeing how often 𐀁 attaches itself to the fish group, including the variety 𐀐. Now if we turn to the Asoka Edicts we find 𐀑 represents the anusvar - and indeed does still in the scripts derived therefrom to this day! It is possible that this is the Brahmin reduction of a sign which in Proto-Indian on grounds of symmetry, was distributed equally about the sign 1. and indeed solus.
c10.s28.11
i.e. 𐀀 𐀁 𐀂 may well be the sign of the nasalisation of the syllable. It has been already observed that the fish group are open syllables. 𐀃 , 𐀄 , 𐀅 are probably also. Nasalisation is often merely euphonic, though of course in some cases a nasalised syllable may make a word quite distinct from the same syllable un-nasalised. It is unnecessary to labour this fact which of course is common to many languages.
c10.s28.12
A comparison of Texts 47 and 48 shows that the sign of nasalisation could be attached (written around) the last syllable or the whole word optionally. Graphically the latter would be analogous to the convention regarding the writing of a pollysyllablic word as an 'integral' compound.
c10.s28.13
In Text 34 𐀆 (𐀇𐀈𐀉𐀊𐀋) is probably for 𐀍𐀎𐀏𐀐𐀑𐀒 See analysis of Table XXVI.