Detail of seal M-847 with inscription: BELTED FISH / CAGED DOT IN FISH / STACKED TWELVE / POT.

I begin the discussion of twelve-stroke Indus signs with a possible numeral, STACKED TWELVE (XII 1).  It always takes the form of four short verticals in a row placed over another two similar rows of four strokes apiece, in other words four over four over four.  Other researchers enumerate this sign as KP142, W201, and Fs F-11.  Wells cites 30 occurrences, 20 from Mohenjo daro, six from Harappa, three from Lothal, and one from Kalibangan.  He also separates out occurrences of doubling of this symbol (W207 with nine occurrences).  Fairservis does not consider it a numeral, but defines it as "rain."  In his view, this is a universal symbol for falling or flowing water. 

Old Chinese variants of yu3, "rain" (Keightley 1985: 217).

However, if "universal" means that everyone uses such this symbol, then Fairservis is not correct.
In Egyptian, a zigzag line repeated three times, one over the other, represents water (Gardiner 1976: 628).  "Rain," hyt, can be spelled out with a twisted cord (h), a man leaning on a staff (hii), two reeds (y), a semi-circular bread loaf (t), and the water determinative (the stacked zigzags).  There is another representation, comprising the symbol for the sky (resembling a coffee table), with a number of vertical lines descending from it, an ideogram or determinative for rain or dew dating to the Eighteenth Dynasty (N4).  An element resembling the Indus COMB may represent rain in the rock art of the American Southwest, but it has a curved line attached to the long stroke that also resembles highly stylized depictions of mountain sheep (Heizer and Baumhoff 1984: 83).  Thus, although it is sometimes identified as "rain," it could just as easily be defined as "sheep."  In early Chinese writing, there is rake-like element with three "tines," as well as a row of dots that descend from the "teeth" of the "rake," sometimes also including a horizontal line that represents the sky (Keightley 1985: 217).  We thus have an example of a symbol made up of dots alone (Indus), lines only (Egyptian N4, possibly American), and a combination of lines and dots (Chinese).  This is less than complete universality in form and does not convince most researchers that the Indus XII 1 represents rain.

Proto-cuneiform often includes a parallel to the Indus sign and this is true of STACKED TWELVE.  The proto-cuneiform "twelve" is written in two ways.  It can be shown as two vertical columns of impressed circles, six in each (N14).  Or it can be represented with two rows of vertical strokes, one row above the other, again six marks in each row (N58).  Three rows do not seem to occur.  It is interesting to note that in Egyptian, "twelve" is not written with twelve strokes.  Instead, a medj (or inverted "U") represents "ten" and two strokes are added for the ones.

Bas-relief tablet K-69 with inscription (right to left): DOUBLE MEN / DOUBLE STACKED TWELVE / BI-QUOTES / RAKE OVER RAKE / DEE / DEE-SLASH (there may be a SLASH beneath the first sign, or this may just be a flaw).

The second Indus sign of twelve strokes (XII 2) is RAKE OVER RAKE, found elsewhere as KP93 and W267.  Wells finds a total of eight occurrences, with just one from Mohenjo daro.  All the other instances appear on a series of duplicate tablets from Kalibangan.  This sign differs from one I term BI-RAKE in the positioning of the upper "rake" element.  In the BI-RAKE, the upper "rake" has its "tines" at the top, as does the "rake" at the bottom.  Combined, they quite resemble a telephone pole with two arms.  In RAKE OVER RAKE, the bottom "rake" has upward-facing "tines," but the upper one has its "tines" facing downward.  This symbol has no exact parallels outside the Indus Valley, although individual "rakes" occur, e.g., in proto-cuneiform and in Southwestern rock art.

Bar seal M-373 with (partial?) inscription: CEE WITH SLASH (?) / DOUBLE DOWN HEARTS / ASTERISK IN DONUT / DOUBLE COMBS (the first sign might be BACK CEE & CEE, with the "slash" just a scratch).

Following Wells, I include in my list DOUBLE COMBS (XII 3).  Wells enumerates this as an independent sign (W283), noting four occurrences.  Fairservis only lists individual signs, the COMB being his P-5, defined as a dative ending, "belonging to," or L-9, "to write, scratch."  Endings are not normally doubled, so his first proposal seems most unlikely.  The second is not much better, since one would not expect to find a verb doubled either.  Koskenniemi and Parpola also take this to be two instances of one symbol (their KP98).

From the Columbia Plateau of North America, a stick figure anthropomorph with "rayed arc"
on or above its head, similar to the Indus COMB (Keyser 1992: 51, fig. 24d).

The individual COMB might depict rain in Southwestern rock art, in America (Heizer and Baumhoff 1984: 83; 306 occurrences in Nevada make it one of the most frequent symbols).  On the northwest coast, though, a similar element has a curve to it and is referred to as a "rayed arc" (Keyser 1992: 61).  This commonly appears over the head of an anthropomorph, less often over a zoomorph.  It is thought to indicate spiritual power, the result of a vision quest.

Blau scraper with scene and inscription from Jamdat Nasr, with
rake-like element behind person on right (Aruz 2003: 39, Pl. 9).

An individual "rake" represents "great" in proto-cuneiform, transcribed GAL.  In some inscriptions, this element is attached to other symbols with a very short "handle," making it more like the Indus COMB.  As an attachment, it may or may not still mean "great."  For example, a six-pronged variety is part of the word SZANDANA, "gardener."  Proto-Elamite also includes variations on a rake-like sign, M038~a including four prongs, M038~b having five, and M038~d sporting seven.  This may not have the same meaning as in proto-cuneiform.  In any case, note that an adjectival element could be doubled, as "great" + "great" would mean "very great" in some languages.

Adinkra symbol duafe, "comb," symbolizing feminine qualities (Willis 1998: 94-5).

An actual comb is also a symbol sometimes, as, for example, in West Africa.  Such an element in Adinkra symbols is duafe, "wooden comb," representing feminine virtues such as patience, prudence, love, and caring (Willis 1998: 94-5). This association of combs with femininity leads to some actual combs being carved in the shape of a woman.  Note that these combs, whether real ones or the symbols, have an extra element that stands up like the "handle" of the Indus RAKE.  Also note that a symbol not closely tied to spoken language, such as Adinkra duafe, is often depicted on an object multiple times.

Detail from seal Blk-2 with inscription: BLANKET WITH 8 TICKS / STRIPED FAT LEG LAMBDA /
BI-QUOTES // LAMBDA / 7-TOED FOOT / TRI-FORK (over unicorn bull).

Today's fourth symbol is BLANKET WITH EIGHT TICKS (XII 4), also known as W531 and Fs G-13.  Fairservis is of two minds as far as the BLANKET is concerned.  On one hand, he suggests that the variations are not meaningful and he subsumes them all under G-13, which he considers a proper name, Mani(y)ettu or Mani(y)ene.  On the other hand, he suggests that the basic SQUARE or RECTANGLE represents a weight, since such items have been discovered in excavations, and the "ticks" are then numerals indicating the weight measure (Fairservis 1992: 100-101).  As I am not convinced that enumeration occurs in the Indus script, I remain unconvinced of this hypothesis.  Be that as it may, Wells finds four occurrences of this symbol, one from Harappa, one from Lothal, and two from Balakot.

Seal M-627 with inscription: FISH / PINCH (?) / GRID (3 X 7) / POTTED ONE /
STRIPED LOOP UNDER FAT CHEVRON / DOUBLE GRIDS (3 X 7).

Among the twelve-stroke signs, there is another checkerboard: GRID (XII 5), also known as W504.  Fairservis does not note all the variations on the GRID, subsuming them under his G-16 which he defines as "collect, store; storehouse or work area."  He notes that this symbol is usually doubled.  However, Wells enumerates the doubled instances separately (W514).  As a single sign, this GRID (3 x 7) appears only once, at Mohenjo daro (M-627).  The following table shows the frequency of each type of GRID, with the second numeral indicating its occurrences when doubled:

GRID TYPE     FREQ.     DOUBLED
3 x 3                  10            5
3 x 4                    8            6
3 x 5                    2            2
2 x 3                    1            5
2 x 4                    1            5
2 x 5                                  2
2 x 7                                  1
TOTAL            22             26

Thus, in a sense, Fairservis is correct that the GRID is "usually" doubled.  That is, there are more occurrences of a doubled symbol than of a single symbol.  But the difference in frequencies is not that great.  Even so, Walls may be correct in separating single instances from doublings, since GRID XII 5 occurs three times in a single inscription, once between a possible PINCH (or POST + SINGLE QUOTE) and a POTTED ONE, and then doubled at the end of the inscription (M-627).

Proto-cuneiform UR3 and variant b1, "roof," note exactly grids but rectangles containing additional strokes.

As a final note on the GRID, I note that sometimes the top portion of the "cult stand" in front of the  unicorn bull has a grid-like form (see M-627 for this also).  This typically has many more subdividing lines than the GRID sign in the inscriptions.  And this part is not grid-like on every stand.  But there might be some relationship between sign and this design element.



Seal L-110 with inscription: SINGLE QUOTE (?) / DOTTED WINDOW WITH ATTACHED POST / DOTS IN OVERLAPPING CIRCLES / QUAD-FORK (?) (note the rough execution of symbols -- perhaps an unfinished seal).

Next, there is the DOTTED WINDOW WITH ATTACHED POST (XII 6), a singleton from Lothal, as Wells notes (his W507).  This symbol bears some resemblance to one that appears on punchmarked coins of later India.  One of these latter symbols is a "boxed cross" or "window," with a small element inside each compartment.  These internal elements are not simple dots or single "quotes," though.  Two are "bowties" and two are in a form known as "taurine head" (Gupta 1960: Pl. II, symbol no. 125).

Two window-like symbols that occur on some Indian punchmarked coins, motifs 124 and 125 (Gupta 1960: Pl. II).

The seventh symbol in this post is FAT EX IN DIAMOND (XII 7).  It occurs elsewhere as KP382, W391, and Fs N-6b.  Fairservis thinks it represents a settlement or town.  Wells notes 14 occurrences, eight of them from Mohenjo daaro, five from Harappa, and one from Lothal.  I see at least one more, from Kalibangan.

Seal M-880 with inscription: STRIPED DOUBLE BELTED AITCH WITH TICK / CUPPED POST /
POTTED ONE / FAT EX IN DIAMOND / VEE IN DIAMOND / POT.

The suggested meaning of settlement or town is apparently a result of comparison with the Egyptian circled "fat ex," O49.  The glyph is a determinative for villages and towns or regions, certainly.  And Mahadevan lumps the FAT EX IN DIAMOND with the CIRCLED FAT EX as well, modifying the suggested meaning to "lower city" as opposed to acropolis (2009).  I am not convinced that the Indus "diamond" is a variation of the "circle," since there is no *TRI-FORK IN DIAMOND to parallel the fairly common CIRCLED TRI-FORK (as well as a corresponding lack of some other symbols that are either unique to diamonds or unique to circles).  In addition, it is terribly risky to use one script to identify symbols in another script that has no demonstrated relationship to the first one.  As we have seen, a "circled ex" or "circled cross" is extremely common around the world, but has very different meanings in different areas.

A different form of "fat ex" or "cross" in a diamond in this detail from a Peruvian poncho (Appleton 1971: Pl. 69). 

The closest analog to the FAT EX IN DIAMOND in proto-cuneiform is BANSZUR, which came to mean "table; container."  This sign is not exactly a diamond, as the middle section is elongated.  Also, the "ex" inside is not an outlined form as in the Indus sign, but a slash with two backslashes meeting it at different spots.  A better parallel is provided by decorative motifs in South American art (Appleton 1971: Pl. 69, Inca).

Seal M-281 with inscription: DIAMOND WITH 8 RAYS / LOOP (?) / TOP / TRIPLE TRIANGLES / POT.

Another rhomboid symbol of the Indus script is DIAMOND WITH EIGHT RAYS (XII 8), also known as KP359 and W393.  Wells notes seven occurrences from Mohenjo daro.  Although the sign has quadripartite symmetry, a feature that is often seen in Native American and African art, I have found no analogs.

Detail from seal H-3 with inscription: CUPPED 3 / GRID (3 X 3) / HAMMER / BI-QUOTES // STACKED 7 / EF PRONGED EXIT / POT (this pairing of STACKED 7 & EF PRONGED EXIT occurs in just over half its occurrences).

The ninth element is more frequent than most of those we have examined today: EF PRONGED EXIT (XII 9), also enumerated KP288, W476, and Fs N-8.  Fairservis suggests that it represents an irrigation sluice, also noting its frequent occurrence following STACKED SEVEN.  Wells notes 22 occurrences, 14 from Mohenjo daro, 6 from Harappa, and 2 from Kalibangan.  I see 40 from Mohenjo daro, 13 from Harappa, 3 from Kalibangan, and an additional one from Lothal, adding up to 57 instances.  Of these, I see 30 alongside STACKED SEVEN.  The numbers are not high enough for statistical analysis, but since more than half of the XII 9 appearances include this pair, it may be significant.  The EF PRONGED EXIT also occurs doubled in a few cases.  This may or may not represent an independent symbol.

Seal M-1170 with inscription: GARDEN / (over) TRI-FORK (perhaps an odd COMB modifying the FORK?).

The final sign for today is GARDEN (XII 10), perhaps actually a COMB aligned horizontally with the "teeth" upward.  It occurs just once at Mohenjo daro (M-1170).  And it appears as an independent symbol only in Wells' list (W285). 

Some of the "teeth" of this odd "comb" do not seem to be simple, straight vertical marks, reminding me of the Egyptian glyph M8, a pool with lotus flowers.  The glyph occurs in sh3, "lotus pool, meadow," and as a phonetic biliteral representing that sound in other words.  The resemblance between the two symbols is admittedly very slight.

From the Columbia Plateau of North America, a motif perhaps depicting men in a canoe, at top (Keyser 1992: 42, fig. 18).

A better analog comes from the rock art of the Columbia Plateau in northwestern America.  Here it may represent men in a canoe (Keyser 1992: 42, fig. 18).

REFERENCES

Appleton, Le Roy H. 1971. American Indian Design and Decoration. New York: Dover (orig. published 1950 by Charles Scribner's Sons, entitled Indian Art of the Americas).

Aruz, Joan. 2003. Art of the First Cities: The Third Millennium B.C. from the Mediterranean to the Indus. New York and New Haven: Metropolitan Museum of Art and Yale University.

Fairservis, Walter A. 1992. The Harappan Civilization and Its Writing: A Model for the Decipherment of the Indus Script. Leiden: E.J. Brill.

Halloran, John A. 2006. Sumerian Lexicon: A Dictionary Guide to the Ancient Sumerian Language. Los Angeles: Logogram Publishing.

Keightley, David N. 1978 and 1985. Sources of Shang History: The Oracle-Bone Inscriptions of Bronze Age China. Berkeley: University of California.

Keyser, James D. 1992. Indian Rock Art of the Columbia Plateau. Seattle: University of Washington.

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

Mahadevan, Iravatham. 2009. "Meluhha and Agastya: Alpha and Omega of the Indus Script" available at: www.harappa.com/arrow/meluhha_and_agastya_2009.pdf

Wells, Bryan. 1998. An Introduction to Indus Writing: A Thesis. Available at: www.collectionscanada.gc.ca/obj/s4/f2/dsk2/ftp03/MQ31309.pdf

Willis, W. Bruce. 1998. The Adinkra Dictionary: A Visual Primer on the Language of Adinkra. Washington, D.C.: The Pyramid Complex.