the fact that the domesticated horse has played an important role in the culture of the Indo-Iranian speakers, and there is no unambiguous evidence for the presence of Equus caballus in South Asia before the second millennium BCE.27)
While various minority languages are very likely to have been spoken in the Greater Indus Valley,28) there appears to have been only one written language. The sign sequences of the Indus texts are uniform throughout their area of distribution in South Asia.
The argument is reinforced by the fact that some of the Indus seals found in the Near East contain typical Indus signs and sequences — this concerns especially the square seals most common in South Asia — while on some other Indus seals — especially the round seals similar to those of the Gulf and Elamite culture, and the cylinder seals of the Mesopotamian type — have common Indus signs but in sequences completely dissimilar from those occurring on native Harappan texts. Statistically, one would expect that the most frequently attested sign (the occurrences of which constitute almost 10% of the Indus texts) would very often be found next to itself, but this is never the case in the Indus Valley. The combination is attested on a round seal probably found in Mesopotamia, which contains only frequently occurring signs of the Indus script, but in unique sequences.
This suggests that Harappans residing in the Near East had adopted the local language(s) which differed from the Indus language. The c u n e i f o r m t e x t s s p e a k n o t o n l y o f a d i s t a n t c o u n t r y c a l l e d M e l u % h % h a , which most scholars identify with the Greater Indus Valley, but also of
27) For the horse, cf. Meadow 1991; Meadow & Patel 1997. For the prehistory of the Aryan languages and their introduction to South Asia, see now Carpelan & Parpola 2001; Parpola 2002a; 2002b (for the Aryan affinity of the D$asa language); 2004 ; 2005; Kochhar 2000; Driem 2001: II, 1070–1103.
28) Cf. Kuiper 1991: 89–96 for a list of 383 “foreign words in the Rigvedic language”; Lubotsky 2001; Parpola 2002a: 92–94; and Witzel 2003  for the original non-Indo- European language of the Bactria and Margiana Archaeological Complex (BMAC); for the Austro-Asiatic, Tibeto-Burman and Burushaski languages, cf. Parpola 1994: 142 and van Driem 1999; 2001: I, 295–297; 421–433; II, 1202f.
〔 47 〕