Monday, September 23, 2013

BOXY INDUS SYMBOLS


Indus seal H-1708 with inscription: BLANKET WITH TICKS & FANCY STOOL / STRIPED MALLET.


In my previous discussion of Indus symbols, I mentioned some square or rectangular signs.  Sign IV 4 is a small square and IV 5 a taller rectangle, which may be variants of a single actual sign.  As noted, some researchers consider these relatively rare signs to be variants of a symbol that can also appear as a diamond (my IV 9) and even as a circle or oval (II 9).  It seems to me that if round shapes and angular shapes were truly equivalent (i.e., variants of one another), there should also be both round and angular variants of other signs that are more common.  For example, CIRCLED FORK (which includes my original circled post or PACMAN III 28, CIRCLED TRI-FORK V 41, CIRCLED SKEWERED CHEVRON V 44, CIRCLED BI-FORK V 45) should have such variants as *FORK IN DIAMOND and *FORK IN RECTANGLE (or *BOXED FORK) (following the practice of historical linguists, I use the asterisk here to indicate a hypothetical form that is not evidenced).  

The table below presents comparisons of rounded and angular forms, noting presence of absence of parallels.  The first inclusion noted (1A) is the QUOTE or DOT; this is found inside circular or oval signs but not in either quadrangular type.  As the reader can see, most signs comprising a round or angular shape with an inclusion or attachment do not have possible variants in all forms.  Highlighted in yellow below is the only series that does appear to have variants across the three possible types: CIRCLED VEE, VEE IN SQUARE, and VEE IN DIAMOND.  The first and last are among the most common Indus signs, while the second is extremely rare.  Highlighted in green is the only other possible example.  Here, though, the round form contains only a FORK, the rectangular “variant” contains a VEE and a very different element (ODD STACKED NINE), and the rhomboid or diamond-shaped version contains a VEE and FORK.  Thus, these three have some similarities but also differences, making for very weak evidence that round and angular shapes are true variants.  This is not conclusive evidence since there is always the possibility that new finds will be made in the future that contain the predicted forms not found in the three volumes of the Corpus.

Insert / Shape
CIRCLE/OVAL
SQUARE/RECTANGLE
DIAMOND
--
II 9
IV 4, IV 5
IV 9
1A inner QUOTE
CIRCLED DOT (III 26), DOUBLE CIRCLED DOTS (VI 59)
--
--
1B inner POST
PACMAN (III 28)
BISECTED RECTANGLE (V 5)
--
1C crossing TICK
--
--
DIAMOND WITH TICK (V 59)
2A inner VEE
CIRCLED VEE (IV 40)
VEE IN SQUARE (VI 13)
VEE IN DIAMOND (VI 14)
2B inner FORK
CIRCLED FORK (V 41, V 44?, V 45, VIII 34)
(VEE & ODD STACKED 9 IN RECTANGLE XV 20)?
VEE & TRI-FORK IN DIAMOND (IX 16)?, TRI-FORK ON VEE IN DIAMOND (X 9)?
2C inner CIRCLE/SQUARE/ DIAMOND
DONUT (IV 38)
RECTANGLE IN RECTANGLE (VII 40)?
--
2D inner CROSS
CIRCLED CROSS (IV 42)
WINDOW (VI 4); ENVELOPE (VI 6)
--
2E inner DOUBLE QUOTES
CIRCLED TWO (IV 41)
DOUBLE BELTED RECTANGLE (VI 7)
--
2F inner POST + DASH
--
TRIPLE BRICK
--
2G attached POST
--
RECTANGLE WITH ATTACHED POST (VI 24)
--
3A inner TRIPLE QUOTES
CIRCLED THREE (V 42)
--
--
3B inner ASTERISK
CARTWHEEL (V 43)
--
--
3C attached FORK
CIRCLE WITH TRI-FORK (V 47)
--
TRI-FORK ON DIAMOND (VII 17)
3D inner LAMBDA (?)
--
--
PENNANT IN DIAMOND (VII 18)
3E inner VEE & SLASH
--
--
VEE & SLASH IN DIAMOND (VII 63)
4A inner TICKS (& DASHES)
--
BLANKET (VIII 4, IX 26, X 2-4, XI 15-17, XII 4, XIV 20)
--
4B protruding RAYS
QUADRUPLE RAYED CIRCLE (VI 66)
--
OCTUPLE RAYED DIAMOND (XII 8)
4C protruding LASHES
CEE WITH LASHES (VI 77)?, DEE WITH LASHES (VI 78)?
--
--
4D inner NET
CROSS HATCHED CIRCLE (VIII 59)
GRID (VII 7, VIII 6-7, IX 25, XII 5)
--
5A inner BISECTED SQUARE
SLASH IN DONUT (V 51); CIRCLED TRIPLE BRICK (VIII 61)?, CIRCLED GRID (IX 50)?, SLASH IN DONUT WITH COMB (X 43)?, CIRCLED COMB & BISECTED RECTANGLE (XII 34)?
RABBIT TEETH IN RECTANGLE (VIII 13)
--
5B inner HAMMER
--
--
HAMMER IN DIAMOND (IX 17)
6A inner CROSS/EX + 4 DOTS
--
DOTTED WINDOW (X 6)
DOTTED EX IN DIAMOND (X 7)
8A inner FAT EX (4 VEES)
CIRCLED FAT EX (X 42)
--
FAT EX IN DIAMOND (XII 7)
8B POT HAT + VEE
LOOP WITH EF PRONGED TAIL (VI  68)?, POT TOPPED CIRCLE (VIII 56)?, CIRCLE WITH TRI-FORK & ? (VIII 57)
--
POT HATTED VEE IN DIAMOND (XII 37, XIV 24)
10A inner RAYED CIRCLE
CIRCLED DIAMOND WITH RAYS (XVI 11)?
--
CARTWHEEL IN FAT EX IN DIAMOND (XVII 4)?

Casting further doubt on the hypothesis that CIRCLE = SQUARE = DIAMOND, the most common rectangular signs are the various types of BLANKET and GRID.  While there is a very rare CROSS HATCHED CIRCLE that might be viewed as a round version of the GRID, there is little or no evidence that either BLANKET or GRID has a rounded or rhomboid variation.  Similarly, while CARTWHEEL is one of the most common round signs, it has no counterpart in either rectangular or rhomboid shape.
Tablet H-1930B with inscription (from right): BOXED QUOTE & DASH / CUP / TWO POSTS.
 Turning now to rectangular signs in the third volume of the Corpus, I find three not previously described.  The simplest is a rectangle containing a single “tick” at the base and a single “dash” attached to one side.  I designate this VI 83, BOXED QUOTE & DASH (it occurs elsewhere only in Wells 2011 as sign number 601).  It is a singleton from Harappa, appearing on H-1930B, a faience tablet with inscription in bas-relief.  The rest of the inscription on this side is CUP / TWO POSTS, a common combination of signs especially on tablets.  Since this reverse of a great many tablets contains the CUP in combination of 2-6 POSTS, it is possible that sign VI 83 is not really a sign at all.  Instead, it may be a very simple icon, i.e., a representation of something, a picture, rather than part of the inscription.  There are certainly a few other tablets that bear an inscription alongside a pictorial element.  On the other hand, this may simply be an unusual variant of the BLANKET, a sign that typically contains variable numbers of “ticks” (short vertical strokes attached to the base and top, and may also contain one or more “dashes” (short horizontals). 

Broken tablet H-1900A (DOTTED BOXED BOXES) & B (2 POSTS).
Another rectangular sign recalls VEE IN SQUARE but contains three subsections rather than just one: XIV 25 DOTTED BOXED BOXES.  It too is a singleton from Harappa, occurring on H-1900B, another faience tablet in bas-relief.  This tablet is broken so the full inscription is not legible.  But a very small remnant of an adjacent element remains next to the break.  Again, it is possible that this is a pictorial element rather than a true sign.  Alternatively, it may be a variant of the RECTANGLE IN RECTANGLE.  For comparison, I show below the broken bar seal M-1367 with a two-sign inscription.  To the right of these two signs is the back end of a short-horned bull fighting another.

Broken bar seal M-1367 with inscription: VEE IN DIAMOND / RECTANGLE IN RECTANGLE
.

 The final angular sign is XIX 4 BLANKET WITH TICKS & FANCY STOOL (shown at the beginning of this post).  It is yet another singleton from Harappa, appearing on H-1708, a bar seal, alongside a STRIPED MALLET.  Both elements comprising this sign are found separately, BLANKET WITH TICKS (8) occurring as XII4 in my list, and the FOOTED STOOL WITH MID EARS/OVAL as IX 21.  Now, XIX 4 does not appear in Fairservis’ list of 230 symbols, but he does not claim to be comprehensive (1992: 149): “Signs not found in this list are either too obscure or are combinations of the list signs and the reader can accordingly arrange them.”  That is, he interprets ligatures (combinations of signs) as conveying the same meaning as a sequence of the same elements.  In this case, XIX 4 BLANKET WITH TICKS & FANCY STOOL = XII 4 BLANKET + IX 21 STOOL. 

An illustration from the Egyptian Book of the Dead, showing the Eye of Horus over glyph O21 (a shrine).

In other writing and symbol systems, rectangular elements frequently appear with varying content.  In Egyptian hieroglyphs, Gardiner’s glyphs O6, O7, O10, O21, and perhaps O36 are such rectangles.  These all represent buildings or parts of buildings.  The first, O6, is a tall rectangle with a “vee” or smaller rectangle in one of the lower corners, representing a rectangular enclosure as seen from above.  It serves as an ideograph in ḥwt, “castle, mansion, temple, tomb.”  A similar rectangle with inner “vee” as well as the small semi-circle (as an independent sign representing bread and/or the sound t), O7 is a variant of O6.  The internal semi-circle represents the sound of the final consonant in ḥwt.  The name of the goddess Nephthys may be written with the latter symbol, O7, with the addition of the larger, reversed semi-circle that represents the bi-consonantal nb (V 30).  As Fairservis would expect for Indus signs, the combined Egyptian glyphs thus convey the same meaning as the sequence nb + t + ḥwt (+ t + snake or seated goddess determinative), that is, Nbt-ḥyt.  The glyph O10 is a similar “ligature,” with an unusually wide ḥwt containing not only the usual “vee” but also a standing hawk (G5, Horus).  Together, these elements convey the name of the goddess Hathor, which may also be written with the simple O6 plus phonetic symbols for the second half of her name (and the snake-on-nb-basket determinative).  The glyph O21 represents the façade of a shrine, serving as ideograph or determinative in sḥ-ntr, “the divine booth.”  As a depiction of a building, this glyph contains a small rectangle for the door and often another stroke or two to delineate other features.  The last of the suggested parallels is O36, a rectangle with six markings around the outside, representing a wall, an ideograph or determinative in inb “wall” and synonyms.  Thus, if Indus ligatures follow the pattern found in Egyptian, Fairservis’ notion is correct and knowledge of the basic signs yields the meaning of complex ones.

Old Chinese versions of qun1 "granary" and kun4 "difficult," characters which in modern script have become rectangles containing very angular versions of the signs shown here.

Chinese writing provides a different model for the interpretation of ligatures.  There is rectangular character, one of the radicals (or basic symbols), which commonly contains another element.  This is the 31st radical, wei2, an enclosure.  Depending on the dictionary, there are 29 (The 5000 Dictionary 1940), 34 (Wieger 1965), or 23 (Far East Chinese-English Dictionary 2000) characters made in this way.  The character ren2 “person” in the box becomes, not ren2 + wei2, but qiu2 “prisoner.”  The character representing a tree, mu4, inside the square is kun4 “difficult.”  And the small square, representing a mouth, kou3, inside the larger one represents hui2 “to return.”  Thus, in Chinese, knowing the basic characters from which a complex character is made does not necessarily help in understanding the meaning of the complex symbol.  If Indus signs follow this pattern, knowing the constituent elements of a ligature may not shed much light on the meaning of the combination.

Blanket designs from Texas rock art (after Newcomb and Kirkland 1967: 187, Pl. 137, 17-C).  The central motif may represent the local version of Tlaloc, the rain god.
In the American Southwest, so-called blanket designs are often found in rock art.  These are quadrangular elements that contain various shapes, including both simple and complex patterns.  These may or may not actually represent blankets and it may be that the different patterns within the borders convey different meanings.  Some “blankets” have extensions from the four corners, perhaps representing tassels.  Some have a head with large eyes protruding from one end, thought by some to represent a rain god (Patterson 1992: 51, citing Schaafsma 1980).  Where there are two heads protruding, the whole may depict “modest scenes of copulation” (Patterson citing Warner 1983).  Interpretation is difficult in this realm since the societies that produced the petroglyphs are often extinct.  But it is likely that there are several meanings conveyed by superficially similar elements.

Depiction of a bag used in ritual at Sanga in Mali (after LeQuellec 2004: 19).
In rock art elsewhere, we may note a few rectangular elements at Valcamonica in the European Alps.  There is a square with a central dot and a rectangle with two horizontally crossing lines resembling the Indus DOUBLE BELTED RECTANGLE VI 7 (Süss 1985: 9, fig. 12).  In a depiction of warriors, one figure is armed with a “post” and carries a BOXED EX that appears to be his shield (1985: 29, fig. 37).  A boulder at Borno bears a “checkerboard” pattern with “fringe” on the right side (1985: 43, fig. 62).  Surrounding petroglyphs include swords, horned quadrupeds, spirals, and other more enigmatic elements.  A more complex representation includes a rectangle topped by a triangle, with varying types of internal and external markings.  These depict buildings, perhaps two-story houses.

Depiction of a ginyu or genie as seen at Sanga, Mali (after LeQuellec 2004: 18-19).
In Africa, rock art also includes rectangular elements at times.  In the Tassili n’Ajjer of the Sahara, a finger-like motif includes rectangular elements inside a larger motif (LeQuellec 2004: 18-19).  The rectangles may be painted different colors or contain various patterns including grids or pictorial elements.  One explanation is that the “fingers” represent clans of the local Fulani people.  Farther south, in the great overhang at Sanga, in Mali, there are a number of more or less rectangular motifs, some resembling the Indus GRID, others bearing more complex motifs (2004: 60-67).  At least some of these depict bags that are worn by participants in a local festival (see esp. fig. 14).  Another element, a “boxed cross” with an angular, bent appendage, represents a ginyu, a supernatural being (2004: 68).  Painted or engraved geometric motifs appear as well, in rock paintings of Nigeria (including one matching the Indus WINDOW p. 76-77) and at Nachifuku in Zambia (an elaborately decorated quadrangular form p. 101).  No attempt is made to interpret these.


REFERENCES
Fairservis, W.A. 1992. The Harappan Civilization and Its Writing: A Model for the Decipherment of the Indus Script. Leiden: E.J. Brill.
Far East Chinese-English Dictionary. 2000. New York: U.S. International Publishing.
Five Thousand Dictionary, The: A Chinese-English Pocket Dictionary and Index to the Character Cards of the College of Chinese Studies. 1940. C.H. Fenn, compiler. Peking: Cambridge University Press.
LeQuellec, J.-L. 2004. Rock Art in Africa: Mythology & Legend. Paris: Flammarion.
Parpola, A., B.M. Pande and P. Koskikallio. 2010. Corpus of Indus Seals and Inscriptions. Volume 3: New material, untraced objects, and collections outside India and Pakistan. Helsinki: Suomalainen Tiedeakatemia.
Patterson, A. 1992. A Field Guide to Rock Art Symbols of the Greater Southwest. Boulder: Johnson Books.
Süss, E. 1985. Le Incisioni Rupestri della Valcamonica. (orig. published 1958) Milan: Edizioni del Milione.
Wells, B.K. 2011. Epigraphic Approaches to Indus Writing. Oxford & Oakville: Oxbow Books.
Wieger, L. 1965. Chinese Characters: Their Origin, Etymology, History, Classification and Signification. New York: Paragon & Dover (reprint of original published in 1915 & 1927 by Catholic Mission Press).

Thursday, September 19, 2013

New Banyan signs and a Man with Cup and Saucer




Shamanic figure holding grain (after Newcomb & Kirkland 1996: 43).
In the months since my last post, I finally obtained the third volume of the Corpus of Indus Seals and Inscriptions (Parpola, Pande & Koskikallio 2010).  As I transcribed the inscriptions in this volume, using the system outlined in this blog, I noted a few symbols I had not seen before (in the first two volumes of the Corpus).  Two of these “new” signs may be considered variations on the DUBYA, a sign with three “stems” joined at the base.  The outer two “stems” curve toward the base, forming a “U” or CUP shape, with the central “stem” vertical.  Many instances of the simple DUBYA seem to occur on pottery, although in virtually all of these cases, they are probably better interpreted as instances of CUPPED POST (over 30 possible occurrences from Mohenjo-daro, Harappa, Lothal, Kalibangan, Banawali, Chanhu-daro, Amri, Chandigarh, and Rahman-dheri). 

 Seal M-172 with inscription: DIAMOND / SHISH KEBAB TOPPED DUBYA / 
FISH / SINGLE QUOTE // STACKED EIGHT / FORK.

Most occurrences of the DUBYA on seals and tablets have some addition to the top of the three “stems” and, usually, all three bear the same top.  These symbols may be related to the variously topped POTS referred to as the BANYAN by Korvink (2008: 28).  Three seals from Mohenjo-daro bear a SHISH KEBAB TOPPED DUBYA (M-172, M-414, M-758).  One doubtful occurrence on a pot shard from Harappa contains a possible FORK TOPPED DUBYA (H-1013).  Another, still more doubtful appearance of a CHEVRON-TOPPED DUBYA may occur on a broken tablet, also from Harappa (H-1339). 

 Tablet H-289, side A (left) with inscription (from right): FOUR QUOTES / 
BUD TOPPED DUBYA // POT; side B (right): STRIPED LEAF.

There are also several instances of a BUD TOPPED DUBYA (H-289, H-290, H-577, K-15).  The first two cited from Harappa appear on tablets, while the third is a seal.  The single instance from Kalibangan is also on a seal.  The majority of the remaining DUBYAS are an apparent simplification of the last type which I previously designated LOOP TOPPED DUBYA (seal M-14, seal H-6, tablets H-300, H-2202 {broken & thus uncertain}, H-901A, H-902A, seal K-63, and probably copper ingot C-40).  There are three different varieties of the BUD TOPPED DUBYA and it makes sense to simply include those topped with loops as a fourth variation if this sign.
Inscription from copper ingot C-40 (from right): STACKED SIX / WHISKERED FISH / LOOP TOPPED DUBYA / STRIPED CIRCLE / FIVE QUOTES / (over) TRIPLE STACKED ROOFS / MAN HOLDING QUOTE / FORK.

In two cases, the each of the three “stems” of the DUBYA bears a different top.  The first of these signs contains eleven strokes, so I tentatively designate it as XI 39 (BUD, SKEWERED CIRCLE & FORK TOPPED DUBYA).  It appears on M-1759.  The second also contains eleven strokes: XI 40 (SHISH KEBAB, FORK & BUD TOPPED DUBYA).  It appears on H-2186.  

Tablet H-2186 with inscription, side A (above, reading from right): SHISH KEBAB, FORK & 
BUD TOPPED DUBYA (?) / FOUR QUOTES / FORK; side B (below): TWO POSTS / CUP.

It is interesting to note that in the second sign, both the SHISH KEBAB and the BUD occur together, clearly differentiated (though my rendering is not very clear).  This strongly suggests that, while the BUD and LOOP TOPPED versions appear to be variants of the same sign, we should not lump these together with the SHISH KEBAB TOPPED DUBYA.  Without the evidence of this single sign, it would be tempting to see the SHISH KEBAB simply as Mohenjo-daro’s regional variant of the BUD/LOOP (which mainly occurs at Harappa, though also found at Kalibangan and Chanhu-daro). 
In fact, this is evidently how A. Parpola viewed these symbols (1994: 72-73).  In his sign list, updated from the earlier KP version (Koskenniemi and Parpola 1982), the DUBYA symbols with tops are grouped together as sign 123.  Variants “b” and “c” have LOOP tops, “d,” “e,” and “f” have BUD tops, and “g”, “h,” and “i” have SHISH KEBAB tops.  Interestingly, Parpola notes the occurrence of the DUBYA with BUD, SKEWERED CIRCLE, & TRI-FORK tops (his sign 113) but not the variation with SHISH KEBAB, FORK & BUD tops.  Had he seen the latter, he might have split the DUBYA signs into two groups.

 Tablet H-2186 with inscription, side A (above, reading from right): SHISH KEBAB, FORK 
& BUD TOPPED DUBYA (?) / FOUR QUOTES / FORK; side B (below): TWO POSTS / CUP.

Another sign occurring once in the third volume of the Corpus was previously noted by Wells (2011:174, sign 115): IX 62, MAN HOLDING CUP & CIRCLE.  This sign appears once, on pot shard H-1506.  It is similar to other signs depicting an anthropomorph holding something.  Most of the time, the figure holds just a single item, but there is a MAN HOLDING DOUBLE DEE-SLASHES (XIII 36) and a MAN HOLDING DOUBLE POSTS (IX 27).  In these two, the signs show bilateral symmetry – they are the same on the right and on the left.  But in IX 62, the two sides do not match.  This may be significant because, when seeking parallels in other areas, symmetrical symbols are much easier to match than asymmetrical ones.

For example, among the Egyptian hieroglyphs, there are quite a few depicting anthropomorphs (men, women, and deities) that hold a single object.  Fewer depict two items being held.  Three examples in Gardiner’s list show a man standing and holding something in each hand, each more or less bilaterally symmetrical (glyphs A37, A38, and A 39).  Two asymmetrical standing figures bear a different object in each hand (A22 and A23).  Each of these holds a staff in the right hand, reminiscent of the Indus MAN HOLDING POST.  The left arm hangs down at the side, though, with the second object held horizontally across the body.  In A22, this second object is the ‘b3 scepter (this scepter being glyph S42); in A23 the second object is a mace (T3).  But none of these Egyptian glyphs is particularly close to the Indus sign.

Chinese characters da4 "big" and jia1 "press."
 
In Chinese, as noted in the much earlier blog concerning the MAN sign, a parallel symbol is the word for “big” (da4).  This Chinese character is a little simpler than the Indus MAN, in modern writing.  The head and upper torso are depicted together as a single vertical stroke which continues downward, curving off to the left side, to form one leg as well.  The arms are a single horizontal stroke crossing the “body.”  And the second leg is another curved line, arising from just beneath the arms and pointing in the opposite direction from the first stroke.  The character also serves as a radical, or base, for forming several more complex characters.  In none of these does the basic figure clearly hold something. 

Old Chinese characters yi4 "and, also" and yeh4 "night."

 But there are a few that are reminiscent of the Indus type which does hold something.  For example, jia1 includes the figure with the widespread arms, beneath which hang two small chevron-like appendages (essentially identical to one form of lai2, “to come,” except that in “come” the central vertical stroke continues down between the spread legs, thus creating the character meaning “tree”).  This forms a character meaning “to press” (among other things).  In this case, the two objects suspended from the arms are identical to each other.  In the older seal writing, there is also a character formed by the person with a dot or “quote” under each arm; this is yi4 which originally referred to sides (now written very differently and used as a conjunction, “and, also”) (Wieger 1965: 158).
Complex Chinese characters are typically asymmetrical but rarely pictorial.  Among those formed on the 37th radical (“big”), few resemble a man holding things.  One that does is qi4, in which the human figure forms the base of the character, with two different elements just above the arms.  The element on the left resembles the Indus SHISH KEBAB (or the Chinese hand) and on the right, a small version of the character for a knife.  Together these elements form the word for “covenant, bond, deed.”

Chinese qi4 "covenant."

In rock art, which is often clearly pictorial, there are many examples of human figures holding various items.  In the American Southwest, for example, a simple anthropomorphic figure may hold an element resembling the Indus SKEWERED DONUT in one hand, which may represent an atlatl or spear-thrower (from Rio Piedra Pintada, Rio Grande County, Colorado, in Slifer 1998: 61).  In the example shown, the other hand grasps a curved stick in one case, while a column of dots extends above the right side of the smaller figure.  A more complex figure, perhaps a shaman, seems to clasp an ear of grain on each side (from Seminole Canyon in Texas, in Newcomb and Kirkland 1996: 43, figure 11.2, detail, shown at beginning of this post).  A third example is from the American northeast, showing an anthropomorph with a bow and arrow in one hand and a possible tomahawk in the other (from an Indian grave marker in Kingston, New York, in Lenik 2002: 167, fig. 133).  

Anthropomorphs bearing objects (after Slifer 1998: 61).

Schematic human figures also appear in some rock art from Africa.  One example shown here seems to depict a warrior mounted on a horse, with a weapon and a shield (LeQuellec 2004: 58, fig. 3, from Kourki in southern Niger).  A second bears two enigmatic staffs (2004: 86, fig. 46, from the Pedra do Feitiço, “stone of the fetish” by the Zaire river in Democratic Republic of Congo).  These are not so very different from dueling warriors depicted on stone at Valcamonica in Europe (Arcà 2004).  Here, two figures face one another, each bearing an apparent weapon in one hand and a shield in the other.  The weapon of the figure on the right is difficult to make out amid the multiple abrasions and pits in the rock, but the “staff” of the figure on the left is quite clear.

Anthropomorph with bow and tomahawk over canoe and dancers (after Lenik 2002: 167).

Thus, while human-like figures appear in places all over the world, the more obscure symbols without symmetry (variously topped DUBYAS) are unique to the Indus Civilization.

African mounted warrior (after LeQuellec 2004: 58).



Anthropomorph bearing weapons (?) after the "Stone of the Fetish" in Congo (LeQuellec 2004: 46).

Duelling warriors from Valcamonica in the European Alps (after Arcà 2004).
REFERENCES
Arcà, A. 2004. “Warriors and Iron Age duels in Valcamonica rock art” in TRACCE Online Rock Art Bulletin 20 (December 5, 2004). Available online at www.rupestre.net/tracce/?p=3809 .
Korvink, M. 2008. The Indus Script: A Positional-Statistical Approach. Gilund Press.
Lenik, E.J. 2002. Picture Rocks: American Indian Rock Art in the Northeast Woodlands. Hanover, NH: University Press of New England.
LeQuellec, J.-L. 2004. Rock Art in Africa: Mythology & Legend. Paris: Flammarion.
Newcomb, W.W., Jr., and F. Kirkland. 1996. The Rock Art of Texas Indians. Austin: University of Texas.
Parpola, A. 1994 & 2009. Deciphering the Indus Script. Cambridge: University Press.
Parpola, A., B.M. Pande and P. Koskikallio. 2010. Corpus of Indus Seals and Inscriptions. Volume 3: New material, untraced objects, and collections outside India and Pakistan. Helsinki: Suomalainen Tiedeakatemia.
Slifer, D. 1998. Signs of Life: Rock Art of the Upper Rio Grande. Santa Fe, NM: Ancient City Press.
Wells, B.K. 2011. Epigraphic Approaches to Indus Writing. Oxford & Oakville: Oxbow Books.
Wieger, L. 1965. Chinese Characters: Their Origin, Etymology, History, Classification and Signification. New York: Paragon & Dover (reprint of original published in 1915 & 1927 by Catholic Mission Press).