Ligatures in the Indus Script: Part 2

Seal M-899 with inscription: SQUARE WY / CHEVRON BEARER / POT BEARER

.

In my previous post, I noted that ligatures of Egyptian phonetic glyphs follow the rule of “simple addition” in significance as well as form: ligature AB = glyph A + glyph B.  For example, Gardiner’s glyph D59 = D58 (leg with foot, representing the consonant b) + D36 (forearm with hand representing the glottal stop ‘).  The ligature represents the combination of these two sounds, ‘b, in the word for “horn” (Gardiner 1976: 458).  When the ligatured glyphs are not phonetic, but determinatives (glyphs that carry semantic meaning but are not pronounced), the meaning of the resulting ligature is not as predictable.  For example, D56 (leg with bent knee, determinative in words for foot, leg, etc.) + T30 (a knife, determinative in words for knife, cut, sharp, etc.) = D57.  The result, still a determinative, appears in words for mutilate, damage, and so on.  More succinctly, in this case, ligature AB = glyph C or AB ≠ A + B. 

Tablet M-495A with inscription (right to left): CIRCLED FORK / CRAB / HAIRY HUNCHBACK / POT / BEARER / TRIPLE CUPS / FORK (note the sequence POT + BEARER, which may be equivalent to ligature POT BEARER).

 In Indus script, terminal signs POT and BEARER occur in sequence POT + BEARER (e.g., K-8, M-209, M-733, L-140, KP 5061) as well as the ligature, POT (HATTED) BEARER.  Thus, this ligature appears to follow the pattern I term “simple addition”: ligature AB = A + B.  But the same pattern is not characteristic of all ligatures.  CHEVRON ligatured to BEARER yields CHEVRON (HATTED) BEARER, but its component parts do not occur in sequence: there is no *CHEVRON / BEARER.  Thus, there is insufficient evidence to conclude that the ligature combines the meanings of CHEVRON and BEARER. 

Seal M-268 with inscription: CARTWHEEL / MAN / CHEVRON / SINGLE POST / STACKED TRIPLE CIRCLES (note the pair CHEVRON + SINGLE POST in this inscription, which may be equivalent to QUOTE UNDER CHEVRON).

Other signs that appear in ligatures with CHEVRON include MAN (BY CHEVRON), FISH, CAGED FISH, FORK, QUOTE (always doubled), THREE POSTS, and EX.  If the ligature CHEVRON + X is equivalent to the sequence CHEVRON / X, we would expect to see some instances of sequences of CHEVRON and one or more of these symbols in inscriptions.  However, CHEVRON usually precedes SINGLE POST (in 14 of 17 inscriptions in KP 1982: 113-114, as well as in Krs-1 and H-2133-2134).  This sequence could be the equivalent of QUOTE UNDER CHEVRON.  But, whereas all instances of QUOTE UNDER CHEVRON are doubled, there is no doubled sequence: that is, *CHEVRON / SINGLE POST / CHEVRON / SINGLE POST does not occur.  CHEVRON does precede EX in one inscription (H-1367), which suggests the ligature EX UNDER CHEVRON may be a case of “simple addition.”  Also, CHEVRON follows THREE QUOTES in two inscriptions and this might be equivalent to THREE POSTS UNDER CHEVRON (3Q + CHEVRON in H-1367 and KP 9903, a Gulf seal).  But there are no inscriptions in which CHEVRON precedes or follows FISH, CAGED FISH, or FORK, the other signs with which it forms ligatures.  It seems, then, that the independent CHEVRON may not be the same element as the “chevron” that appears in ligatures, despite their graphic similarity.

Table 1 below lists the signs that appear in ligatures with CHEVRON, comparing these with sequences of the same two signs:

Sign A	Sign B	Ligature AB	Sequence A + B	Sequence B + A
CHEVRON	MAN	MAN BY CHEVRON	--	--
CHEVRON	FISH	FISH UNDER CHEVRON	--	--
CHEVRON	CAGED FISH	CAGED FISH UNDER CHEVRON	--	--
CHEVRON	FORK	FORK UNDER CHEVRON	--	--
CHEVRON	QUOTE	QUOTE UNDER CHEVRON	CHEVRON + 1 POST (14)	--
CHEVRON	EX	EX UNDER CHEVRON	CHEVRON + EX (1)	--
CHEVRON	3 POSTS	3 POSTS UNDER CHEVRON	--	3 QUOTES + CHEVRON (2)

 Table 1. Ligatures and sequences with CHEVRON.

In the previous post I noted that ligatures of MAN (a terminal) and another sign (where the anthropomorph holds an item) are not terminals.  The difference in function suggests that the “simple addition” rule does not apply to these ligatures.  Similarly, ligatures of POT (a terminal) and a “numeral” (where one to four "quotes" appear inside the POT) are not terminals, once more failing to confirm a hypothesis of “simple addition” ruling ligatures.  While some of the evidence for ligatures with CHEVRON could conceivably be interpreted as supporting this hypothesis, for most of these ligatures the absence of a matching sequence suggests the hypothesis is not confirmed. 

Seal M-650 with inscription: CIRCLED VEE / BI-QUOTES / DOUBLE QUOTES UNDER CHEVRONS / RAKE / FISH / SPEAR (note that QUOTE UNDER CHEVRON is always doubled, which is unlike the sequence POST + CHEVRON).

A more common medial symbol that appears in a number of ligatures is the FORK.  A miniature FORK is attached to the following signs: SPACESHIP, THREE POSTS, FAT EX, FOOTED STOOL, VEE IN DIAMOND, BI-RAKE, CIRCLE, and OVERLAPPING CIRCLES, as well as in MAN HOLDING FORK, CUPPED FORK, FORK & VEE IN DIAMOND, CIRCLED FORK, and FORK TOPPED BATTERY).  Since FORK is one of the most common medial signs, it seems reasonable to expect it in combination with at least some of the signs with which it co-occurs in ligatures.  Table 2 compares ligatures containing FORK with sequences:

Sign A	Sign B	Ligature AB	Sequence A + B	Sequence B + A
FORK	SPACESHIP	SPACESHIP W/ ATT. FORK	--	--
FORK	FAT EX	FAT EX W/ ATT. FORK	--	--
FORK	FOOTED STOOL	FOOTED STOOL W/ ATT. FORK	--	STOOL + FORK (1)?
FORK	BI-RAKE	BI-RAKE W/ ATT/ FORK	--	--
FORK	BATTERY	FORK TOPPED BATTERY	--	BATTERY + FORK (1); MALLET + FORK (72)?
FORK	3 POSTS	3 POSTS W/ ATT. FORK	--	3 QUOTES + FORK (20)
FORK	OVERLAPPING CIRCLES	OVERLAPPING CIRCLES W/ ATT. FORK	--	OVERLAPPING CIRCLES + FORK (2)
FORK	MAN	MAN HOLDING FORK	FORK + MAN (4)	--
FORK	CIRCLE	CIRCLED FORK; CIRCLE W/ ATT. FORK	FORK + CIRCLE (3?)	CIRCLE + FORK (1)
FORK	VEE IN DIAMOND	FORK & VEE IN DIAMOND; VEE IN DIAMOND W/ ATT. FORK	FORK + VEE IN DIAMOND (1); FORK + CIRCLED VEE (2)	--
FORK	CUP	CUPPED FORK	CUP + FORK (2)	--

Table 2. Comparison of ligatures of FORK with sequences (number of inscriptions in parentheses).

There is no sequence to compare with the ligature for FORK and SPACESHIP, FAT EX and FORK, FOOTED STOOL and FORK (though STOOL + FORK appears in H-773), or BI-RAKE and FORK. 

FORK follows THREE QUOTES in multiple inscriptions (these are the only signs in H-585, M-179, K-65, H-1664, K-97, M-1857; the sequence also occurs in H-239, M-251, M-95, M-318, Ns-7, K-82, M-292, M-1364, H-54, H-70, H-735, H-462, L-79, and KP 8726).  If we assume that THREE QUOTES = THREE POSTS, the sequences of THREE QUOTES + FORK tend to confirm the ligature THREE POSTS WITH ATTACHED FORK as equivalent to the sequence.  However, FORK also follows other apparent numerals, including TWO POSTS (x 10), FOUR QUOTES (x 27), FIVE QUOTES (x 10), STACKED SIX (x 8), STACKED SEVEN (x 6), STACKED EIGHT (x 3), and STACKED NINE (x 1).  Curiously, there are no ligatures of FORK with any of these other “numerals.”  This leads to the question: If ligatures are abbreviations of common sequences, why is there no *FOUR QUOTES/POSTS WITH ATTACHED FORK?

Seal M-409 with inscription: CIRCLED FORK / POT / 3 POSTS WITH ATTACHED FORK.

The same question arises upon examination of other data.  FORK follows OVERLAPPING CIRCLES in two inscriptions (KP 2754 and M-411); but it follows SLASHES IN OVERLAPPING CIRCLES 17 times (M-1795, M-115, C-1, M-7, H-659, KP 2145, M-907, M-507 & -508, M-1457 & -1295, M-1330, M-1953, M-166, KP 5056, M-391, L-110, M-125, M-1889, M-1807).  It is curious, then, that there is no ligature *SLASHES IN OVERLAPPING CIRCLES WITH ATTACHED FORK.  At best, then, there is limited evidence that OVERLAPPING CIRCLES + FORK = OVERLAPPING CIRCLES WITH ATTACHED FORK.  But consider the whole inscription of M-605 where this ligature appears: on side A of this copper tablet there is the sequence THREE QUOTES / FISH / MALLET / FORK / POT; on side B there is only the one sign, OVERLAPPING CIRCLES WITH ATTACHED FORK.  If ligatures are abbreviations, it seems rather odd that the ligature/abbreviation appears on the least crowded side of this tablet.

Tablet M-605B: OVERLAPPING CIRCLES WITH ATTACHED FORK (note there is only the one sign on this side of the tablet, which calls into question the interpretation of ligatures as abbreviations of sign sequences).

FORK precedes MAN in 4 inscriptions (KP 6001A, M-1770, M-1918, H-19).  But FORK also precedes other terminals, including (POT x 61, BEARER x 5, CHEVRON BEARER x 2, POT BEARER x 1).  There is no ligature of FORK with any of these other terminals.  This leads to another question: why should there be a ligature of FORK + MAN, when both sequence and ligature are rare, whereas FORK + POT, an extremely common sequence, is never ligatured?  In addition, there is the usual change in function; whereas MAN is a terminal, the ligature MAN HOLDING FORK is medial.  Again, the question of the purpose of ligatures arises.  If a symbol was desired to abbreviate a common sequence, we would expect to see *POT WITH ATTACHED FORK or perhaps *POTTED FORK, neither of which occurs.

Tablet M-582A with inscription (right to left): CIRCLED FORK / (SLASHES IN) OVERLAPPING CIRCLES / REN LAMBDA / FORK / CIRCLE / CEE / POT (note the sequence FORK + CIRCLE which may be equivalent to CIRCLE WITH ATTACHED FORK; also note that the other comparable ligature, CIRCLED FORK appears here as well).

 FORK precedes CIRCLE in a few inscriptions, mostly duplicate tablets (M-582 to -588, -1535 to -1443, -2048, -2050, -1563, -2054, -601 to 604; possibly in M-724).  CIRCLE precedes FORK as well (C-22 and M-131).  These sequences seem to provide limited evidence supporting the “simple addition” hypothesis for CIRCLE WITH ATTACHED FORK.  But is that what this suggests?  Or is this evidence for CIRCLED FORK?  For that matter, is the singleton CIRCLE WITH ATTACHED FORK the equivalent of the common CIRCLED FORK?  The former occurs only in M-1154, where it appears at the end of the inscription, after the terminal SPEAR.  In this position, the ligature could conceivably be a terminal sign.  But note that CIRCLED FORK appears in the same inscription.  Thus, these two ligatures, though they combine similar elements, cannot be equivalent. 

Seal M-1154 with inscription: AITCH (?) / SINGLE QUOTE (?) / CIRCLED FORK / DOT IN FISH / SPEAR / CIRCLE WITH ATTACHED FORK (note the co-occurrence of CIRCLED FORK & CIRCLE W/ ATT. FORK shows these two ligatures, although similar, differ fundamentally).

And what are we to make of the co-occurrence of FORK and CIRCLED FORK in the same inscription?  In M-77, CIRCLED FORK immediately precedes FORK; in 35 other inscriptions the two signs are not contiguous.  Since FORK can occur in multiples (three FORKS in M-2094 and KP 6001A, four in M-2093 and M-1123), the co-occurrence of FORK with the ligature CIRCLED FORK does not disprove the hypothesis of “simple addition.”  More significant is the co-occurrence of the proposed ligature, CIRCLED FORK, with the sequence FORK + CIRCLE on the many duplicate tablets.  This indicates that CIRCLED FORK cannot be the equivalent of the sequence.  

Broken seal L-46 beginning with the sequence FORK + VEE IN DIAMOND.

 

Similarly, FORK & VEE IN DIAMOND resembles VEE IN DIAMOND WITH ATTACHED FORK: the same elements are ligatured in both cases, though in different ways.  Are they equivalent?  And is one or both equivalent to the sequence FORK + VEE IN DIAMOND (one inscription, L-46) and/or FORK + CIRCLED VEE (two inscriptions, M-314 and K-78)?  Again, there are also two inscriptions containing FORK & VEE IN DIAMOND followed immediately by FORK (H-511 and M-918).  As before, since FORK can appear in multiples, this does not negate the possibility of “simple addition” in the ligature.  But it seems odd.

Detail from seal M-1116 with inscription: BED / RAYED DIAMOND (?) / POTTED ONE / SINGLE QUOTE / STRIPED LEAF (?) // (2nd row) CORN HOLDER / STRIPED BOWTIE / FORK & VEE IN DIAMOND (the final sign may be equivalent to the rare sequence FORK + VEE IN DIAMOND or FORK + CIRCLED VEE).

The same oddity appears in M-784 and KP 2408 where FORK follows CUPPED FORK.  The sequence of CUP and FORK does also occur: L-101 contains CUP + FORK, M-495 contains TRIPLE CUPS + FORK.  So this may support the hypothesis of “simple addition.”  However, the last inscription (M-495, see illustration above) is unusually long, with 26 signs appearing in inscriptions on each side of the prism-shaped tablet.  If the purpose of a ligature is to include more information in a smaller space (i.e., an abbreviation), this seems like the type of inscription that most needs a ligature.  The three CUPS and the FORK appear squeezed unusually thin, with the “tines” of the “fork” obscured due to crowding.  Nevertheless, the person who made this object did not put the FORK inside any of the CUPS to save space.  So saving space, i.e. abbreviating, seems not to have been the purpose of ligatures.  Perhaps, then, ligatures serve a different function from sequences composed of the same (or similar) elements.

Detail of seal M-649 showing inscription: FORK TOPPED BATTERY / CARTWHEEL.

There is only one inscription pairing FORK with BATTERY (H-689: BATTERY / FORK / POTTED THREE).  This is very meager evidence supporting FORK TOPPED BATTERY being a ligature of these two signs.  The pair MALLET + FORK is far more frequent, though (73 inscriptions).  It would make more sense for there to be an abbreviation of this common sequence.  So perhaps the FORK TOPPED BATTERY should be re-analyzed as FORK TOPPED MALLET.  Examining the inscriptions for evidence on this point, I note that FORK TOPPED BATTERY occurs in initial position in all three inscriptions where it appears (M-1263, M-2012, M-649).  In comparison, the sequence MALLET + FORK precedes BI-QUOTES as part of the prefix only once (Lh-1).  Usually MALLET + FORK either comprises the medial segment or is the last part of that segment before the terminal (62 inscriptions, plus three others in which the terminal POT is followed by the prefix constant SINGLE QUOTE and one or more additional signs).  Interestingly, in M-998, the sequence MALLET + FORK precedes CARTWHEEL, providing an exact parallel to the ligatured FORK TOPPED BATTERY + CARTWHEEL of M-649.  All this is interesting, but with so few parallel examples no conclusion is certain.

Detail of seal Lh-1 with inscription: MAN HOLDING DEE-SLASH / MALLET / FORK / BI-QUOTES / FISH UNDER CHEVRON / FOOT / MAN BY CHEVRON / GRID / FORK TOPPED POT / POT (if MALLET + FORK is equivalent to the ligature FORK TOPPED BATTERY, one would expect the ligature rather than the sequence in this long inscription).

I will continue the discussion of ligatures in the next post.  But thus far, the evidence for the hypothesis of “simple addition” does not seem adequate to explain the occurrence of ligatures. Sequences of signs generally do not appear to be equivalent to the same signs combined in ligatures.  Ligatures occur where there is plenty of room to include the signs in sequence and sequences occur where crowding would seem to indicate the need for a ligature.  Thus, my tentative conclusion is that apparent ligatures should not be interpreted as simply the sum of their parts.

REFERENCES

Gardiner, Sir A. 1976 (1927). Egyptian Grammar: Being an Introduction to the Study of Hieroglyphs. Oxford: Ashmolean Museum and Griffith Institute.

Joshi, J.P. and A. Parpola. 1987. Corpus of Indus Seals and Inscriptions. 1. Collections in India. Helsinki: Suomalainen Tiedeakatemia.

 

Koskenniemi, K. and Parpola, A. 1982. A Concordance to the Texts in the Indus Script. Helsinki: Department of Asian and African Studies, University of Helsinki.

Parpola, A., B.M. Pande, and P. Koskikallio. 2010. Corpus of Indus Seals and Inscriptions. 3. New Material, Untraced Objects, and Collections Outside India and Pakistan. Helsinki: Suomalainen Tiedeakatemia.

Shah, S.G.M. and A. Parpola. 1991. Corpus of Indus Seals and Inscriptions. 2. Collections in Pakistan. Helsinki: Suomalainen Tiedeakatemia.