PRIMITIVE NUMBER SYSTEMS. 585 



they oount as liigii as 10 by means of tlieir lingers. * Teu they exi)iess 

 by theii- Avord for '"all." The Bushnieu have the same number limit, 

 expressing- any number greater than 2 by the (Miuivalent word for 

 ''many." The Veddas of Ceylon count ''ekkamai,'' 1, "dekkamai," 2, 

 and then continue by repeating again and again tlie word "otameekai," 

 meaning '';ind one niore."t 



Numerous as are the instances in which two stan<ls as the number 

 limit for savage tribes, three is thus used still more frequently. The 

 New Hollanders have no names for numbers greater than tliree.^ The 

 low forest tribes of Brazil comiuonly express any number greater than 

 three by their equivalent for "w/^n<</."§ The Australians of Herbert 

 River do the same.|| The Fuegans are supposed to have counted for- 

 merly to ten, but at the present time tlieir entire number system is com- 

 prised in the tiiree words : "kaoncli,'" 1; "compaipi,"' 1'; "maten,"3.** 

 TheCampas, of Peru, count: "patrio,"!; "pittem," 2; '^mahuimi," 3. 

 Beyond this they can express no number except by some such expres- 

 sions as 1 and •>, I nnd 1 and 3, etc., showing the total indicated by hold- 

 ing up the i)r()per number of Angers. As a definite number anything 

 beyond ten is to them inconceivable, and tliey refer to it as "to haiue," 

 "many.-'tt The Australian tribe of the Wiraduroi have no numerals 

 which enable them to count beyond 3. With them four is "many," and 

 five "very many." Almost exactly the same statement may be made of 

 the Dip]>il, the Kamilaroi, the Adelaide, the Turrubnl, the West Aus- 

 tralia, the Encounter Bay, the New South Wales, and the Tasmania 

 tribes.i:| Some of these indicate four by expressions such as "two-two," 

 or "two i)air," and live by '• two-three," or " two-two-one." The Encoun- 

 ter Bay tribe nses an analogous reduplication for six even, saying 

 "kuko-kuko-kuko," that is, "two- two-two." The Yaucos, of the Ama- 

 zon, express the mimber three by tlie astounding word " poettarraror- 

 incoaroac," at which La ( 'ondemaine duly remarks : §§ "Happily for those 

 who have dealings with them, their arithmetic goes no further." 



The general limitation of the number sense existing among the low 

 races of the world now begins to become apparent. Specific words exist 

 for one, two, three, etc., and beyond that anything is " numy." The en- 

 tire number system of a tribe may be ''one," " many," or it may be " one, 

 two, many." More numerous yet are the cases when^ the counting goes 

 one step farther, and gives "one, two, three, many," as the s<'.ale througii 

 wiiicli th''. savages' number sense can conduct him. In the same way 



■ Miillcr: Grundriss der Sprachicissenschaft. li. iv, p. 47. 



t Dechamp's L'Anthropolofjic, 1S91, p. .318. 



t Tylor Primiiive Citllurc, vol. i, p. 213. 



\S Op. r(/.,p. 242. 



II Liiiiiboltz, C, IlitHcliii (If hi Sociti' d' Aiithvo])o!i>ii'n' tlv I'uris 



*-" Op. cii., 1.S87, p. :mo. 



ft Weiiicr. I'croii tl Holiric, p. 5(»0. 



tt Miillcr, Orutidrina der ■•^j))'., \i. I\'., Abteiiung 1, miillu lorn. 



^^ Voyage de la Uirierc dts Amazons, p. (j4. 



