You don't even need emergent properties. It's the fundamental insight of Darwinism that the statistical properties of individuals, as revealed, necessarily in aggregate, are of the first importance in modern scientific explanation. This was carried over into physics by Maxwell, Gibbs, and ultimately the quantum revolution, and of course is the basic principle of statistical reserach in the social sciences. I'm not saying taht they aren't emergent properties. I think there are. But there are statistical ones based onm mere agregation before you even get to that. So, how can you test statistical generalizations? Let me refer to to some good methods books, Bill. As usual, you're blowing smoke. jks  Miles Jackson <cqmv@pdx.edu> wrote: <BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid"> On Tue, 28 Jan 2003 billbartlett@dodo.com.au wrote: > If your analysis is only applicable to abstract "large groups", but not > to actual real people, then how can we test its veracity? It is just a > vague and sweeping generalisation. > > Bill Bartlett > Bracknell Tas > This assumes that "large groups" cannot have emergent properties that are distinct from its constituent parts. Understanding how one computer server works does not provide a useful understanding of the Internet (the emergent whole). We need to pay attention to the unit of analysis here. If our goal is to understand "social facts", the test of any hypothesis requires an observation of the social ensemble, not simply the individual elements that make up the ensemble. --Even further: an accurate theory of the ensemble may appear completely wrong when applied to a specific, individual case. That doesn't make the theory any less accurate or useful. Miles </BLOCKQUOTE> <hr size=1>Do you Yahoo!? <a href="http://rd.yahoo.com/mail/mailsig/*http://mailplus.yahoo.com">Yahoo! Mail Plus</a> - Powerful. Affordable. <a href="http://rd.yahoo.com/mail/mailsig/*http://mailplus.yahoo.com">Sign up now</a>