On the other hand, there are also various versions of methodological holism. The general common denominator for these views is that social phenomena are not reducible to mere explanations of their constituent members. For sociologist Émile Durkheim, for instance, there exist 'social facts' that go beyond merely individualistic explanations: some phenomena can only be described at the level of structures and systems (social, legal, ecological, demographic, architectural, historic, genealogical, religious, economic, geographic, etc.). This implies the possibility of introducing sociological interventions to try to remedy large-scale social problems that may seem otherwise intractable. These interventions may appear downright insane and counterintuitive sometimes (like installing blue lights to reduce crime and suicide rates, or painting prison walls pink to reduce violence and rapes), but that's not to say they don't work. I hate to link to Cracked.com, but since we're not doing scholarly research right now... fuck it.
Unless you have some training in this type of thing, methodological holism may seem at first glance like some vague notion referring to equally nebulous concepts unsuitable for rigorous scientific research and investigation, but once you see it in action in the fascinating introduction to sociology by Nicholas Christakis below, you'll probably end up wanting to major in sociology. :)