An amazing result from work with Piraha tribe in the Amazon has shed some light on the importance of language in human cognition. You can read a summary on the Science web site, here:
The Piraha have a counting system that goes "one, two, many." This lack of higher quantitation is pretty darn rare, and it may explain why they can't properly work with or conceptualize numbers greater than two. There are some dramatic experiments in the study, including one where Piraha are asked to duplicate a number of stripes written on a piece of paper (there's a picture of that in the story). When the number goes above two, they can't do it. In the example shown, they just keep drawing three stripes, even when there are more than that in the original.
The basic idea is that if you don't have a way to describe something, you can't think about it. Thus, people don't have an inherent number sense, and if a human lacks a word for a number, they can't operate well with it.
2004-08-20 04:26 pm UTC (link) DeleteFreezeScreen Select
The Golden Ratio (Mario Livio) has a brief section about how evidence points to initial counting systems all using the "one, two, many" system. In addition to evidence that birds are unable to distinguish more than four objects that suggests a biological basis, Livio also suggests linguistic evidence. Here's an excerpt:
"More pieces of evidence suggest that the initial counting systems followed the "one, two, ... many" philosophy. These come from linguistic differences in the treatment of plurals and of fractions. In Hebrew, for example, ther eis a special form of plural for some pairs of identical items (e.g. hands, feet) or for words representing objects that contain two identical parts (e.g. pants, eyeglasses, scissors) that is different from the normal plural. Thus, while normal plurals in in "im" (for items considered masculine) or "ot" (for feminine items), the plural form for eyes, breasts, and so on, or the words for objects with two identical parts, end in "ayim." Similar forms exist in Finnish and used to exist (until medieval times) in Czech. Even more important, the transition to fractions, which surely required a higher degree of familiarity with numbers, is characterized by a marked linguistic difference in the names of fractions other than a half. In Indo-European languages, and even in some that are not (e.g. Hungarian and Hebrew) the names for the fractions "one-third", "one fifth", and so on generally derive from the names of the numbers of whihc these fractions are reciprocals (three, five, etc.). In Hebrew, for example, the number "three" is "shalosh" and "one-third" is "shlish." In Hungarian "three" is "Harom" and "one-third" is "Harmad." This is not true, however, for the number "half," which is not related to "two." In Romanian, for example, "two" is "doi" and "half" is "jumate"; in Hebrew "two" is "shtayim" and "half" is "hetsi"; in Hungarian "two" is "ketto" and "half" is "fel." The implication may be that while the number 1/2 was understood relatively early, the notion and comprehension of other fractions as reciprocals (namely "one over") of integer numbers probably developed only after counting passed the "three is a crowd" barrier.