So, if you don't know who Latour is, what you really need to know is that his primary theory, using the terminology that's been largely settled on (although he's still not entirely happy with it) is a thing called "actor-network theory," or ANT for short. The jist of it is that we tend to look at the world (and particularly science) in terms of subjects and objects. Subjects do things, and objects have things done to them. Pretty much textbook grammar, right? Except that Latour argues that not only is that a very limited view of the world, it's flat out wrong. It's not subject and object, it's human and non-human; they're not one prioritized above the other in terms of intent or action or ability, it's everything acting on everything else. Non-humans are capable of acting on their environments and on humans, just as humans do the same. Non-humans mediate human relationships and humans mediate the relationships between things. It's not a top-down hierarchy, it's a great sprawling network of connections and relationships and the in-between spaces of connections where things actually happen.
Also, non-human things aren't just things, like tables and computers (already a pretty complex series of connections and networks), but are also things like organizations and concepts and events. The history and practice of science, according to Latour, isn't just what happens in a lab report, but also the lab in the context of the organization, and the personal networks of the scientists who live there, and the various groups or people who have stakes in the research, and the places the materials for the research come from, and the political situation that surrounds the lab funding -- all of that comprises science and its practices, not just a given experiment in a given lab at a given time.
There is a lot more to his ideas, of course, including the idea that since humans make non-human objects, those objects contain scripting that tell humans how they should interact with the non-human objects in question. Design and bias, conscious or not, pervade the things we do and make, even if we don't think about it most of the time. He argues for the reclamation of human relationships with our surroundings, abandoning a false nature/social split that predicts the decline of the world and our corruption from a pre-historical cultural and moral purity in favor of embracing that human and non-human actors/actants are actually very closely entwined and will only become more so, and that this is a good thing as we realize that this has pretty much always been the state of man, rather than some form of degeneration (he's not a big fan of postmodernism, as this may show).
Whew. That's a lot of stuff and I've barely scratched the surface. Suffice it to say that Latour is a figure of controversy and has been for a number of years in all sorts of disciplines.
For me personally, I think what he's saying makes a great deal of sense. I'm not heavily invested in top-down vertical power structures, though, even if it's just people and things we're talking about. My whole life is networked effects, so it seems intuitive to me. More to the point, though, it works with my research. I think that the way we use things is heavily wrapped up in encoded social and cultural expectations. Looking back to the 18th century and the advent of consumer culture in Britain (as a sort of environmental/historical petri dish to set up our experiments in), it's a great way for us to see how non-human actants complicate questions of class and political visibility.
Even better, though, are objects in fiction. Why, you may ask? Because the 18th-century saw the advent of realistic fiction, where objects started being called out and included purposefully rather than just as a brief sketch tied into a plot event, mentioned once and never again. Objects weren't included as stage setting with no purpose in fiction in this period and setting -- if you as an author included blue curtains, for example, there were some real reasons why those curtains were blue, and it was related to the story or the character in question, and you could be pretty darn sure that your audience (largely homogenous culturally) would get it (or get it explained to them).
So if literary objects (the objects authors put in their novels) consist largely of references to real objects and cultural truths that humans are expected to understand, then they aren't anything but embedded cultural scripts. And if they're used to illustrate or embody cultural boundaries of race, gender, class, status, etc.... then how they are used (whether as intended or subverted) through their relationship with the human characters is pretty darn telling as to what those scripts might be, which tells us a lot in turn about society and the expectations (often unstated) that exist in that time and place. That in turn can teach us how to look at our own time and place and things and relationships, and maybe draw some conclusions for ourselves that are a bit overdue.
So yeah, that's Latour, at least from my perspective.
*Works that inform this include: "The Berlin Key," Pandora's Hope, We Have Never Been Modern, Reassembling the Social