Working With Critical Security Studies in Real Research
Critical Security Studies doesn't hand you a ready-made toolkit. When you sit down to actually apply it, the first problem you hit is that there are at least five competing sub-traditions within the field, and they don't always agree with each other. Copenhagen School, Welsh School, Paris School—they all use the word "security" differently, and if you pull quotes from one without acknowledging the others, your work falls apart under review. I spent months trying to merge securitization theory with practices-based approaches for a paper on migration borders. The problem was structural. Securitization treats security as a speech act—you declare something a threat and suddenly normal politics is suspended. Practices-based approaches treat security as something you do repeatedly until it becomes routine. These aren't incompatible, but they're not obviously compatible either. I had to build a bridge by treating speech acts as themselves a kind of practice—something that only works when embedded in institutional routines. That's not in any textbook. You figure that out by getting stuck.
Navigating Critical Security Studies And World Politics
The most useful thing to understand early on is that CSS is not a method. It's a stance toward the concept of security itself. You start by assuming that whatever counts as a security issue was not always treated that way, and might not always be treated that way. Climate change is a recent addition to the security domain. Cyber warfare isn't either. Human security as a framework barely existed before the 1990s. These aren't natural categories. They're political achievements. Here's how I actually go about a CSS analysis. First, pick a concrete security claim—a policy document, a speech, a legal framework. Something that declares or implies a threat. Then trace who is authorized to make that declaration and who is excluded from that authorization. A state minister declaring a border crisis has institutional power behind the speech act. A displaced person describing the same border has none. The gap between those two positions is where the analysis lives. Next, map what gets rendered invisible by the securitization. When a government frames drug trafficking as a national security threat, human rights abuses at the border often drop out of the policy discussion entirely. The threat construct does work—it pulls resources and attention toward itself. But it also silences other concerns. My advice is to literally keep a running list of what the security framing pushes aside. Without that list, you're just restating the official position in slightly more critical language.
One thing beginners consistently get wrong is treatingCSS as purely deconstructive. It's easy to spend an entire paper dismantling a security claim and then have nothing left to say. The field actually expects you to do something after the critique. Barry Buzan and Ole Wæver wrote about this explicitly—their model of securitization includes the possibility of dem securitization, where a threat is pulled back into normal politics. The Paris School scholars like Didier Bigo push this further by asking what alternatives to securitization actually look like in practice. You should be able to answer that question, or at least point toward one. Another trap is the objectivity fallacy. CSS researchers often position themselves as uncovering hidden power structures, but the researcher's own positioning is itself a security-relevant act. When you study border security, your access to certain documents, your institutional affiliation, your language proficiency—these all shape what you can see and what you cannot. I learned this the hard way when I spent three months researching a counter-terrorism policy and realized my university affiliation made certain officials willing to talk to me while others refused. That wasn't a methodological nuisance. It was data. The access pattern itself told me something about how the security community reproduces itself.
Get the Full Details
Common Pitfalls and What Actually Works
The biggest bottleneck in CSS research is empirical grounding. The theory is strong on critique but weak on specifying what would count as evidence against it. If someone tells you your securitization analysis is wrong, it's not always clear what finding would convince them. This is partly by design—the field is skeptical of positivist standards—but it's also a genuine weakness. Papers that simply refuse to engage with counter-evidence tend to be dismissed by mainstream IR regardless of how internally consistent they are. A workaround I've found useful is to anchor your analysis to a specific institutional moment with a clear paper trail. Budget allocations, legislative debates, classified document leaks, internal memos that become public. These give you something concrete to track. Instead of arguing generally that a threat was constructed, you can show the moment when the funding shifted, the language changed, the responsible agency was created. That's harder to dismiss as mere interpretation. The field also struggles with scale. Most CSS work focuses on the state level or the local level. There's not much in the way of tools for analyzing security at the planetary scale without either collapsing back into state-centric analysis or drifting into vague cosmopolitanism. I've found that borrowing from political ecology helps here—concepts like regime complexity and multi-scalar governance give you vocabulary for talking about how security operates across levels without pretending there's a single authoritative frame.
If you're reading CSS for the first time, start with the basics and then immediately read the critiques. Security: A New Framework for Anarchy by Buzan, Wæver, and de Wilde is the obvious starting point. Then read Rob Biorio's work on the ontological politics of securitization, which questions whether the speech act model actually captures how threats work in practice. Then move to Didier Bigo's notion of the "ban-opticon" and Stuart Croft's work on the mundane practices of security. You'll develop a much more useful sense of the field's actual shape than if you read only the foundational texts. The practical reality of doing CSS work is that it rarely changes policy directly. Policymakers don't cite securitization theory in their briefing documents. What it does is change how you read the documents they do produce. It makes the familiar strange. A counter-terrorism strategy becomes an artifact of a particular historical moment rather than a rational response to an obvious threat. That's the value. It's not a policy manual. It's a way of seeing that tends to persist even when you stop actively practicing it.
