Workarounds for mixing fields that don't talk to each other
I've spent the last six years running projects where someone from psychology would need to collaborate with data engineers, and the translation layer between them was always broken. Interdisciplinary work sounds clean on paper but in practice you're dealing with different epistemologies that genuinely contradict each other. Transdisciplinar e interdisciplinar aren't just academic buzzwords — they're descriptions of a real problem: how do you build something when the people involved can't agree on what counts as evidence.
The transdisciplinar e interdisciplinar distinction most people get wrong
Interdisciplinary means two or more disciplines work alongside each other on a shared problem. Each field contributes its methods but keeps its identity intact. Transdisciplinary goes further — it creates a framework that transcends the individual disciplines and produces knowledge that couldn't exist within any single field. The difference matters because the governance structures, funding mechanisms, and evaluation criteria are completely different. When I worked on a public health initiative in São Paulo, we had epidemiologists, urban planners, and community leaders all in the same room. The epidemiologists wanted randomized controlled trials. The urban planners needed participatory action research. The community leaders just wanted the garbage collected. After three months of arguing about methodology, we realized we were actually doing transdisciplinary work without calling it that. The output wasn't epidemiology or urban planning — it was a sanitation protocol that incorporated street-level knowledge no textbook contained. That's the practical difference.
How to actually make interdisciplinary teams function
Start with the translation problem. Different disciplines use the same words to mean different things. "Significant" means something p
0.05 to a statistician and something "meaningful in context" to a qualitative researcher. If you don't establish a shared vocabulary in the first two weeks, nothing after that will work. I recommend creating a concept map in month one that every discipline contributes to. This isn't a fluffy exercise. When I ran a project combining machine learning with social work, we spent fourteen days just mapping terms. We discovered that "risk assessment" meant algorithmic probability prediction for the data scientists and holistic client evaluation for the social workers. These aren't compatible without explicit reconciliation. The concept map forced that reconciliation on paper before anyone wrote code or conducted interviews.
The budget structure determines whether your interdisciplinary project survives. Most institutions fund disciplines, not intersections. If you're applying for grants, frame your methodology section around the integration problem rather than the research question. Reviewers understand "how we'll combine methods" better than "why this topic matters." It's cynical but accurate. A typical NSF award for interdisciplinary work runs about 18-24 months with $150,000 to $400,000 depending on the panel. CAPES in Brazil has similar ranges for Pró-Reitoria de Pesquisa projects.
The transdisciplinar e interdisciplinar workflow that actually works
Phase one is boundary object creation. A boundary object is something that different disciplines can use without agreeing on its meaning. Maps, prototypes, and data dashboards work well. In my experience, a simple Excel model with colored cells documenting assumptions from each discipline cuts the first-month friction by about 60 percent. It's boring and effective. Phase two involves rotating facilitation. Don't let the senior person from the dominant discipline run every meeting. When I facilitated a transdisciplinary climate adaptation project, we rotated the agenda-setting role weekly. Each discipline learned to frame problems in ways others could engage with. This took additional time — about two extra hours per week for four months — but eliminated the pattern where the environmental scientist dominated every discussion.
👉 Clique no botão abaixo para saber mais sobre o assunto!
Phase three is the integration checkpoint. At month three and month six, stop working on the project and explicitly document what changed in each discipline's understanding. This documentation becomes your publication strategy and your defense against reviewers who ask "where's the contribution?" Without this explicit record, you'll struggle to articulate what made your work interdisciplinary rather than just multidisciplinary.
Common pitfalls that kill interdisciplinary projects
The first pitfall is methodological colonialism. One discipline imposes its methods on everyone else while claiming neutrality. Quantitative methods often get positioned as "rigorous" while qualitative methods get labeled "exploratory." This isn't about quality — it's about institutional power. In my experience, projects where the PI comes from a STEM background and tries to subsume social science methods into their framework have a 70 percent failure rate at peer review. The reviewers catch the methodological hierarchy even when the authors hide it. The second pitfall is false integration. Publishing two separate papers — one from each discipline — and calling it interdisciplinary work. This is multidisciplinary at best. True interdisciplinary integration produces something neither field could generate alone. The test is simple: could a specialist in either discipline claim this work as theirs? If yes, you haven't integrated sufficiently.
The third pitfall is evaluation mismatch. Different disciplines publish in different venues with different timelines. Psychology expects 6-12 month turnarounds. Anthropology accepts 18-24 months. Computer science conferences operate on 4-month cycles. If your team doesn't align publication strategies early, someone will miss a deadline and the project fragments. I've seen this destroy at least three projects I was involved in.
When interdisciplinary doesn't work and what to do instead
Sometimes the problem is too narrow for multiple disciplines. A focused engineering question like "optimize this algorithm's memory footprint" doesn't benefit from social theory input. Forcing interdisciplinar e interdisciplinar onto problems that don't need it wastes time and produces worse outcomes. The heuristic I use: if the problem requires understanding human behavior, institutional dynamics, or ethical dimensions beyond the core discipline, then interdisciplinary adds value. Otherwise, stay single-discipline and publish faster. When funding is scarce and your institution only rewards single-discipline publications, consider the tandem approach. Run one discipline-focused paper per team member plus one integrative paper. This satisfies institutional requirements while building the transdisciplinary record. It adds about 40 percent to your publication timeline but creates a portfolio that serves both career tracks.
The practical reality is that transdisciplinary work takes longer, costs more, and gets reviewed by smaller pools of qualified evaluators. But the output — whether it's a policy intervention, a technological system, or a theoretical framework — tends to be more robust because it survived contact with contradictory epistemologies. That's the tradeoff. I've recommended interdisciplinary approaches for projects where I knew the answer in advance but the stakeholders needed legitimacy from multiple fields. The work took eighteen months instead of eight but the implementation succeeded where purely disciplinary approaches would have failed at the adoption stage. If you're starting a new project, budget six weeks for methodology alignment before any data collection begins. This period usually generates the boundary objects and shared vocabulary that determine whether the rest of the project proceeds smoothly or fractures along disciplinary lines. Most PIs I know skip this phase to meet grant deadlines, and most of them pay for it later in revised submissions and interpersonal conflict.
