That’s all well and good in practice, but how do humans work in theory?

In our public policy theory class, we began with a discussion about the absence of an overarching theory of public policy. I learned pretty quickly that other social science disciplines–economics perhaps excluded–lack one as well. Why? Three potential reasons:

1. We tend to understand things atomistically/essentially, rather than relationally. Whereas I intuitively understand human behavior as a product of the influences around a person, we seem still to understand individuals through the traits and characteristics by which we divide and categorize them. How can we possibly understand a person’s behavior if we don’t know anything about the people that influence them, and the people that influence those people?
2. We tend to focus on cause and effect as the process by which we understand phenomena. This may be hard-wired into humans (my brother would say something here about Kant and Hume), and it may have led to marvelously explanatory statistical models, but it seems impotent in the face of what to me are some of the most interesting questions: What does it mean when people say a party has energy, or doesn’t? What turns a protest into a mob? When does dinner become a date? What makes some communities decide to welcome immigrants, where others reject them? Why does a legislature convulse around a particular topic when they do? Causal chains may not be the most effective mechanism for understanding these phenomena that are central to our experience. The example that a dancer gave to me recently was of the first few minutes of her dance class, which consisted of an exercise designed entirely to get the group in sync with each other. After just a few minutes, the group began to move as one; understanding the causal chain there simply won’t teach us as much as looking at the system as a whole.
3. We tend to look at text, rather than subtext. Humans are constantly picking up non-verbal cues from others and their environment; how these get processed (sub)consciously turn around and launch similarly murky processes in others; this is what I think of when people talk about “emergence.”

What am I missing? What am I wrong about? What should I read next?