SITE MAP Home    Indian Lexicon     Corpus of Inscriptions    Artefacts   Decipherment

Decipherment of  pictorials

Back to Table of Contents

A start is made with the decipherment of the field symbols (pictorials) in inscriptions.

Out of about 3000 inscriptions recorded in the corpuses, over 1990 inscriptions contain ‘field symbols’. An object may have many sides and each side may feature a different field symbol with or without accompanying inscriptions (Mahadevan, p. 9; Table VIII: Distribution of field symbols by sites, pp. 777-779). Thus, field symbols constitute the major message component of the corpus of inscriptions.

‘Reading’ these pictorials is an imperative to successfully interpret the ‘underlying language’ and ‘meaning’ of the inscriptions.

There will be continuing disagreements on the ‘orthographic values’ to be assigned to some ‘images’ or pictorials. For example, is the ‘unicorn’ an imaginary construction of a single horn on an ox or did such an animal, in fact, exist? Is the ‘image’ of a ‘bat’ a ligatured fish? If the fish sign a variant of a ‘loop’? There are, however, many images, which are emphatically deciphered orthographically, for example in such motifs as, elephant, tiger, bison, rhinoceros. A start can be made for the decipherment effort with such images with emphatic, orthographic clarity. Such a beginning will provide valuable clues to categorize the life-activities connoted by the texts and pictorials in inscriptions contained in the inscriptions.

Many pictorials are mostly calligraphically definitive. For example, a bull is a bull is a bull; an elephant is an elephant is an elephant. Many signs can also be interpreted as derived hieroglyphics derived from pictorial symbols.

There are, of course, problems in interpreting the orthography of some signs, for exampe:
Does this sign represent a fish or a loop?
Is this sign depicting a jar or a the face of a bull?
Is this sign depicting a circle, an axle or a wheel with six spokes?

Hypotheses

Two interlinked hypotheses govern this study for script interpretation and may be elaborated further, as follows:

Hypothesis 1: Indian languages are derived from the lingua franca of the Sarasvati-Sindhu civilization. Selected lexemes of Indian languages provide the morphemes required to attach ‘sound-bites’ to the pictorials in inscriptions of the Harappan script, thus attesting to the continuity of the civilization in the present-day spoken languages of the sub-continent.

Hypothesis 2: The pictorials in inscriptions of the script represented ‘meaningful’ messages related to the life-activities of the civilization and, these messages can be read from ‘homonyms’ of the morphemes attached to the pictorials in inscriptions (cf. Hypothesis 1).
 

All words are semantic indicators. ella_ccollum porul. kur-ittan-ave_ (Tol. Col. Peya. 1)

Through a number of monographs, superb structural analyses of the inscriptions have been done by both Parpola and Mahadevan. The analyses point to the use of most of the signs as representing ‘nouns’.

Use of rebus method

Rebus (Latin: by means of things) is a graphemic expression of the phonetic shape of a word or syllable. Rebus uses words pronounced alike (homophones) but with different meanings. Sumerian script was phonetized using the rebus principle. So were the Egyptian hieroglyphs based on the rebus principle.

The use of the rebus methodology is justified on the following evidence and analysis:

According to the Parpola concordance which contains a corpus of 2942 inscriptions, 300 inscriptions are composed of either one sign or two signs. Many signs occur in predictable pairs; 57 pairwise combinations account for a total frequency of 3154 occurrences (32% of 9798 occurrences of all pairwise combinations). Given the statistical evidence that the average length of a text is 5 signs, it is apparent that one sign or a pair of signs represents a ‘substantive category’ of information, i.e., a complete message.

In addition to the field symbol, the texts of the inscriptions are composed of an average of five signs. The longest inscription has 26 signs (found on two identical three-sided tablets: M-494 and M-495 of Parpola corpus).

There are over 170 inscriptions with only one sign (in addition to the field symbol); about 30 inscriptions have only two signs (Sepo Koskenniemi et al., 1973, p. x)

A number of signs appear in duplicated pairs: for example,
Sign 245 occurs in 70 pairs.
(nine squares in a rectangle or a chequered-rectangle)

These are apparently not duplicated alphabets or syllables.

Many pictorials in inscriptions in field symbols also occur in pairs: two tigers, two bisons, two heads of the unicorn.

These statistics establish the following facts:
 

• A combination of pictorials without the use of any sign constitute the message.
• One or two signs and/or a pair of signs are adequate to compose the core of the messages.

This leads to the apparent conclusion that the solus sign or each sign in a pairwise combinations (which constitute the core of information conveyed) is not an alphabet or a syllable, but a WORD.

This apparent evidence is echoed in Koskenniemi et al: "... the Indus script is in all likelihood a relatively crude morphemographic writing system. The graphemes would usually stand for he lexical morphemes... This hypothesis is based on the approximate date this writing system was created (circa 26th century B.C.), the parallel presented by the Sumerian writing system of that time (the Fara texts of the 26th century), the brevity of recurring combinations, and the number of different graphemes." (Koskenniemi and Parpola, 1982, pp. 10-11). Another echo is found in the structural analysis of Mahadevan: " G.R. Hunter (1934, p. 126) formulated a set of criteria for segmentation of the texts and found that almost every sign of common occurrence functioned as a single word. The Soviet group (M.A. Probst and A.M. Kondratov in Y.V. Knorozov et al., Proto-Indica, Moscow, 1965) analyzed texts on the computer and concluded that the Indus script is essentially morphemic in character, resembling the Egyptian hieroglyphic system in this respect. I have described the logical word-division procedures developed by me (I. Mahadevan, "Recent advances in the study of the Indus script", Puratattva, Vol. 9, p. 34), which show that most of the signs of the Indus script are word-signs... no one has so far been able to establish by objective analytical procedures the existence of purely phonetic syllabic signs in the Indus script... Phonograms formed by the rebus principle can be recognized only if the underlying language is known or assumed as a working hypothesis. Since the identity of the Harappan language has not yet been established beyond doubt, I cannot be said that any phonogram has been recognized with certainty... It is however very likely that there are rebus-based phonograms in the Indus script, as otherwise, it is very difficult to account for the presence of such unlikely objects such as the fish, birds, animals and insects in what are most probably names and titles on the seal-texts. It is likely that the Indus scrip resembles in this respect the Egyptian scrip in which pictographic signs serve as phonetic signs based on the rebus principle (e.g. the picture of a ‘goose’ stands for ‘son’ as the two words are homonymous in the Egyptian language). It is no always possible in the present state of our knowledge to distinguish between ideograms and phonograms..." (I. Mahadevan, "Towards a grammar of the Indus texts: ‘intelligible to the eye, if not to the ears’, Tamil Civilization, Vol. 4, Nos. 3 and 4, Tanjore, 1966, pp. 18-19).

Orthography and analysis of some sequences of graphemes in the inscriptions
 
Parpola notes (1994, pp. 84-85), echoing similar observations by Mahadevan:
"…a few signs are indeed found mostly at the end of inscriptions, notably sign 342 (‘jar’ grapheme)and sign 211 (‘arrow’ grapheme) and they are major aids in the segmentation of texts. The sign 342 (‘jar’ grapheme) is by far the most common sign of the Indus script, representing about 10 percent of all sign occurrences. About one-third of all inscriptions end with this sign…the sign is never found at the beginning of inscriptions…The sequence sign 102 (‘three short strokes’) followed by sign 192 mainly occurs at the end of inscriptions, and is never followed by the usual ‘end’ sign 342 (‘jar’ grapheme)…"

The sign 192 appears to be a four-fold ligaturing of the underlying basic sign: 
 

Parpola notes (1994, pp.103-104): "A comparative study of the allographs provides one important means of identifying the iconic meaning of even fairly abstract shapes…the (allograph) continuum)…Taken together, these signs can be understood as pictures of a single object, namely, ‘steps, staircase or ladder’; taken individually, such a conclusion would hardly be possible."
 
 On the use of circumgraphs associated with the 'fish' sign, Parpola notes (1994, pp.69-70): "…the four strokes around the ‘fish’ sign may in fact be understood to be read after it, and that their meaning is close to the sign ‘arrow’ that is often found in this position." The following sequences are shown as evidence.

5477
1554

4604
5477

Twenty signs occur with the circumgraph of four short strokes; many of these 20 signs occur as final motifs of the text, functioning similar to the ‘jar’ sign which terminates many texts. The circumgraph may, therefore, be the terminating ‘word’ of the text, functioning similar to the 'arrow' sign. The 'arrow' sign terminates 184 inscriptions (out of a total of 227 inscriptions in which the 'arrow' occurs).

Back to Table of Contents