The take off point for several emergence working group discussions were phenomena of human social/cultural change that follow technological or policy changes not intended to bring them about. The explosion of social networking on the web  is one (of many) examples of the former. A collection of stories of the latter (again, not at all comprehensive) is provided by Stephen Dubner and Steven D. Levitt in a NY Times article titled "Freakonomics: Unintended Consequences ."
We can all, of course, add our own personal examples of unintended consequences, stories of times when we took a particular action with a particular intent in mind and found ourselves, for better or for worse, being a cause of things we hadn't imagined would result. We might chalk those up to having not carefully enough thought through what we were doing, and promise ourselves to do better next time. And we might ask technology developers and policy makers to promise the same thing, or get angry at them (or ourselves) for not having done enough homework.
There is, though, another possibility, the possibility that human social organization is itself an "emergent system", one in which it is not even in principle possible to fully predict the future outcomes of a particular action, no matter how well anyone has done their homework. Systems consisting of quite simple elements interacting in quite simple ways can have the property that the outcome from a particular starting point cannot be determined except by trying it out (cf Conway's Game of Life) . It is not then at all inconceivable that the same thing could hold for the much more complicated elements and interactions involved in human social behavior. Let's hold that thought for a minute.
The full title of the book I was reminded of is "The Nature of Science: The "Problem of Unconceived Alternatives and its Significance ." In it, the author argues from the history of science that scientific theories have always been replaced by previously "unconceived alternatives" (quantum physics being one example among many), and asserts there is no reason to expect that this process will not continue indefinitely. A reviewer of the book expressed concern with the idea that there are always "unconceived alternatives" in the following terms
"... if a scientific community's repertoire of concepts expands over time as it interacts with the world, then that community cannot generate theoretical knowledge effectively"
Let's hold that thought for a minute too.
"Pure chance, only chance, absolute but blind liberty is at the root of the prodigious edifice that is evolution" wrote Jacques Monod in 1970, an idea more recently developed by Stephen J. Gould: "When we fail to accurately predict, that is just nature's reality." Arguably the most elaborate and successful organization that has ever come into existence, life itself appears to be an emergent system , one that has always been and continues to be full both of significant consequences and of alternatives as yet unexplored. Indeed one might characterize biological evolution as an ongoing exploration of both unknown consequences and yet to be materialized alternatives. And one might in turn identify an interesting and productive relation between consequences and alternatives. Current evidence suggests, for example, that high levels of oxygen in the earth's atmosphere is the result of its production by early life forms. That almost certainly had consequences, probably including the extinction of large numbers of organisms for whom high levels of oxygen are lethal. And it created alternatives that had not been explored up to then: organisms (including our ancestors) for whom high levels of oxygen are essential. Consequences create alternatives? So, maybe both consequences nobody anticipated and alternatives no one conceived predate humans and we can/should learn to live with them, perhaps even value them?
What the preceding suggests is that one really ought not be surprised by consequences no one intended nor by alternatives no one conceived. We are ourselves almost certainly the products of such consequences and alternatives, and the processes that gave rise to us are continuing all the time not only all around us but within us. Our bodies, including our brains, are constantly acting and dependent on consequences no one intends (ourselves included) and exploring resulting possibilities no one (again ourselves included) conceives. Why then should we expect all consequences to be "intended" or all alternatives "conceived"?
The answer seems to have to do with the "bipartite" or "hybrid" character of the human brain . As a consequence of our evolutionary past, we have retained the ability to emergently respond to consequences and generate alternatives but have acquired in addition to that something quite different: an ability to imagine both consequences and alternatives that have not yet existed and to make use of those to create "stories" that in turn influence our behavior. To put it differently, we live and act not in one world but in two (at least): in one, the one we share with all other living organisms, there are consequences and alternatives that we deal with (usually quite effectively) without cataloguing (or even being aware of) either. In the other, our story world, we intend consequences and conceive alternatives. And we tend to presume that either we, or someone else, or something else (natural or supernatural) actually knows or could in principle figure out what is going on. It is in this story world, and this story world alone, that there exist "unintended" consequences and "unconceived" alternatives.
Is it useful to try and predict consequences and to conceive new alternatives in our story worlds? Of course. It gives us the ability to avoid problems that we might not otherwise avoid and bring into existence things that might not otherwise have come into existence. Sometimes. And that "sometimes" is a key point. We can certainly try to conceive all alternatives that we can conceive of, and avoid all consequences that we can imagine that aren't ones we intend. But to believe that this will eliminate "unconceived" alternatives or "unintended consequences" is to mistake our story world for the world we actually live in. And that in turn can have quite serious unintended consequences, including rendering one blind to alternatives yet to be conceived. Our ability to create stories gives us quite valuable new capabilities to participate in the "ongoing exploration of both unknown consequences and yet to be materialized alternatives". We should value it for that, and avoid the temptation to think it frees us from the uncertainties and insecurities of an ongoing emergent process that we derive from and in which we necessarily continue to exist. It is, in fact, the uncertainties and insecurities that give us the capability to ourselves influence that process, to create "meaning" where none would otherwise exist . Life, from this perspective, is not an examination in which one is graded on one's ability to conceive all alternatives and avoid unintended consequences. It is an ongoing exploration of new consequences and new alternatives.
If science specifically, and inquiry generally, is understood as an effort to move closer and closer to a definitive description of "reality" in terms of an underlying set of "properties and rules ", one in which everything is predictable and there would therefore no longer be unintended consequences, then indeed unconceived alternatives would be fatal to the effective generation of "theoretical knowledge". One can, however, see both science and inquiry in a quite different light, as an explorer of consequences from which arises as yet unconceived alternatives. From this perspective, unconceived alternatives are not in fact a problem for theoretical knowledge, they are instead its objective. Science and inquiry can proceed perfectly well in the face of unconceived alternatives , just as they always have. But we could better recognize that the most important contributions are not those that show a particular existing alternative way to make sense of things to be better than other existing alternatives. The most important contributions are those that bring into existence previously unconceived alternatives. Science and inquiry too are part of the ongoing exploration of consequences and alternatives.
What really interests me about all of this right at the moment is its implications for thinking about human social organization, both theoretically and practically. An immediate implication is that human social organization should indeed display emergent characteristics, ie the interactions of humans ought to have consequences and yield new alternatives. Some of these should be "predictable" to any given observer/story teller and others not. This seems to me to be obviously the case, as is the corollary that different observers/story tellers may differ in what they regard as predictable and what not. More interesting (to me at least) is the question of whether human social behavior is "just" an emergent system . The issue arises because humans are themselves story tellers, ie each functions in part in relation to intended (as well as unintended) consequences and conceived (as well as unconceived) alternatives, both of which are likely to be not only somewhat different in different people but to be themselves variable and influenced by both other people and cultural practices. Hence human social organization is not only an emergent system. It, like the human brain, is a hybrid one  in which factors lacking a plan or intent interact with others that involve plans and intents. I know of no explicit theoretical inquiries into hybrid systems defined this way, but my guess is they would display (as human societies seem to) properties and characteristics not evident in purely emergent systems including, probably, a more significant influence of top down processes  and an enhanced influence of imagined worlds. Whatever sense trying to account for things in terms of underlying properties and rules might or might not make in the natural science, it would seem to be even less so in the social sciences. And even more so that the productive task would be instead to come up with new as yet unconceived alternate ways of understanding things.
This in turn suggests some significant practical considerations. Just as scientists/inquirers need to develop ways to proceed that replaces movement towards "reality" with an aspiration to bring into existence previously unconceived alternatives, so too it seems to me that those interested in producing social change need to develop ways to proceed that replace movement towards any fixed utopia with a a commitment to bringing into existence previously unconceived alternatives. Social communities, after all, can and will "self-organize" without leaders. That's not to say that individuals shouldn't aspire to trying to facilitate change in particular directions, but it is to say that anyone aspiring to do so needs to be aware of self-organizing tendencies, the roles that diverse individual stories play in them, and the reciprocal interdependences of individual and community stories. It is not enough to know more about the factors at play than anyone else does, nor enough to be an effective "mananger", to be skilled at process. Nor is it enough to be familiar with "best practices". What is additionally required is to be what I have elsewhere called a "fuschia dot ", a master story teller/reviser who is willing and able to assure the kind of continuing story exchange and revision out of which previously unconceived stories emerge. Social organization too is part of an ongoing exploration of consequences and alternatives .
"So this is why Paul and I are bookends: Paul is always interested in expanding the universe of possible explanations (generativity) and I am always looking for ways to narrow down the number of candidate explanations" ... Mark Kuperberg, Emergence Group Listserv, 26 March 2008
"or rather than bookends, a loop between expansion and contraction" ... Anne Dalke, Emergence Group Listserv, 27 March 2008
Perhaps a loop in which contraction serves the function of creating previously unconceived alternatives?