Interdisciplinary IS Research: Opportunity or Identity Crisis?

Every IS paper I have read this year cites a theory from somewhere else. Transaction cost theory from economics. Structuration theory from sociology. Protection motivation theory from psychology. Resource-based view from strategic management. The list goes on. At some point during my comps preparation I tried to find an IS paper that rested entirely on IS-native theory and came up short for longer than I expected. The field borrows relentlessly from its neighbors.

This is sometimes described as a strength, and I think it genuinely is. The problems IS studies, how technology and organizations shape each other, do not respect disciplinary boundaries. A question like "why do employees work around enterprise systems instead of using them as designed?" requires organizational theory, cognitive theory, and enough understanding of the technology to know what the workaround actually is. No single discipline owns all of that. Borrowing is not a sign of weakness. It is what interdisciplinary research looks like when it is working.

But Benbasat and Zmud (2003) identified something real when they named the IS identity crisis. Their argument is that the field drifts from its core as papers increasingly use the IT artifact as background scenery rather than the central object of inquiry. If you can remove the technology from the model and the main argument still holds, you have written a paper about organizational behavior or psychology that happens to mention software. You have not written an IS paper. The IT artifact, meaning the hardware, software, data, and the practices that form around them, needs to be theorized directly, not gestured at. Benbasat and Zmud published this in MIS Quarterly in 2003, and the field had another round of hand-wringing about it, and then largely continued as before.

Sarker and colleagues (2019) actually measured the problem rather than just diagnosing it. Their analysis of papers in major IS journals found that a majority were what they call Type I: papers where the IT role is nominal, meaning the technology is present but not theorized. The artifact gets mentioned, maybe even measured, but the causal logic does not run through it in any meaningful way. My recollection from the study-hub notes is that roughly 56% of reviewed IS papers fell into this category. Only a small fraction achieved the kind of substantive IT role that Benbasat and Zmud had called for. If those figures are accurate (and they are specific enough that I believe them), the identity crisis is not a historical artifact. It is present tense.

The publishing challenge this creates is genuinely strange. IS papers that borrow heavily from sociology risk looking like mediocre sociology to a sociologist reviewer. The IS framing will seem lightweight, the engagement with the sociological literature will feel incomplete, and the treatment of the technology will be a foreign object. At the same time, a paper that engages deeply with sociological theory and writes for an IS audience may get reviewed as "too theoretical" or "not enough IS contribution." You can get caught in the middle: not sociological enough for sociology journals, not applied enough for IS journals, and not sure which reviewer panel to trust.

I have watched this happen in real time. A colleague submitted a paper that drew on institutional theory from organizational sociology in a fairly sophisticated way. One reviewer said the IS contribution was not clear. Another reviewer from the same journal said the institutional theory framing was not rigorous enough. Both reviewers were right in their own terms, and the combination of their feedback was almost impossible to satisfy without rewriting the paper as two different papers for two different audiences.

The opportunity side of this is also real, though. IS has a genuine claim to problems that no single parent discipline can address alone. Algorithmic accountability requires understanding both the technical architecture of machine learning systems and the organizational and institutional structures that govern their deployment. No one discipline has all of that. IS, at its best, is positioned to integrate those perspectives in a way that produces insight neither side would reach alone. The papers I find most interesting right now are the ones that do exactly this without being embarrassed about the borrowing. They take a theoretical lens seriously, engage with it carefully, and use it to illuminate something about how technology and organization interact that would have been invisible without it.

The practical question for an IS PhD student is how to position work that borrows heavily from another discipline. My working answer is that the IT artifact has to be central, not peripheral. If the theoretical lens comes from sociology, the argument still has to run through what the technology does, how it changes what is possible, how it enables or constrains the social dynamics you are studying. If you remove the technology and the argument collapses, you are writing IS research. If you remove it and the argument survives, you are borrowing the IS context to publish sociology. Reviewers can usually tell the difference.

What I find genuinely productive about this debate is that it forces IS researchers to be precise about what they are contributing. A paper that says "I am using TAM plus two new constructs" is making a different kind of claim than one that says "I am showing that the nature of the IT artifact in agentic systems changes the causal logic of adoption research." The second claim requires you to understand both the technology and the theory well enough to explain why the interaction between them produces something novel. That is a harder standard, but it is the one worth aiming at.