7. Literature Synthesis as Boundary Work

Synthesis is claim-making, not stacking

A literature review summarizes; a synthesis makes a claim. The synthesis paper that adds value to a transdisciplinary field is one that says something no single source said — and could be wrong, falsifiable in principle.

The discipline is the same as in primary research. If your synthesis can't be challenged because it doesn't make any claim, it isn't yet a synthesis. The point is to take a position on how the evidence fits together.

Critical appraisal across paradigms

A common mistake: applying a single appraisal tool (CASP, AMSTAR, PRISMA-S) to studies from multiple paradigms. These tools were designed for studies within their home paradigms. A randomized trial and a phenomenological study cannot be evaluated by the same checklist; the rigor concepts differ.

Paradigm-appropriate appraisal:

• Positivist / post-positivist quantitative studies — internal validity, external validity, construct validity, statistical conclusion validity. Tools: CASP RCT checklist, AMSTAR for systematic reviews.
• Interpretivist qualitative studies — trustworthiness criteria (credibility, transferability, dependability, confirmability). Tools: CASP Qualitative, COREQ, SRQR.
• Mixed-methods studies — integration quality, meta-inference quality, plus paradigm-specific criteria for each strand. Tool: MMAT.
• Critical / participatory studies — catalytic validity (does the study advance emancipation?), reciprocity, accountability. Less standardized; document case-specifically.

The transdisciplinary reviewer applies the right tool to each study. Doing so requires you to know each study's paradigm before you appraise it.

Synthesizing as integration, not comparison

Comparing findings side-by-side is preliminary work. Synthesis goes further: it proposes an integrative claim that no single source could make alone.

A useful pattern. Take two findings from different paradigms on the same topic and ask:

• Where do they converge? (Same finding from different methods strengthens it.)
• Where do they diverge? (Divergence may indicate the methods are measuring different facets.)
• Where do they complement? (One reaches what the other can't; together they cover more.)

The synthesis paragraph names all three relations and proposes how they fit. If everything is convergence, you're probably under-reading divergence. If everything is divergence, you're probably missing the layer at which the methods agree.

Writing reviews that build bridges

A review that builds bridges does several things at the level of writing:

• Names the position it's taking, rather than presenting findings as if they speak for themselves
• Honors multiple paradigms rather than implicitly privileging one
• Marks where disciplinary boundaries created blind spots and what evidence was found that the dominant paradigm missed
• Identifies what the integrated picture suggests for both research and practice
• Acknowledges its own limits — what the review couldn't see, what it didn't search

The verbs matter. A review that uses "demonstrates," "shows," "establishes" is a review hiding its interpretive moves. A review that uses "we argue," "we read these findings as," "this synthesis suggests" is owning the position.

Identifying gaps: methodological, paradigmatic, structural

When you identify a "gap," classify it:

• Methodological gap — no one has used the right method (longitudinal, mixed, multi-site). Calls for new methods.
• Paradigmatic gap — no one has studied this from a particular paradigm. Calls for a paradigm shift.
• Structural gap — the community in question has been systematically excluded from study, often because of historical exploitation or current refusal. Calls for partnership, ethical engagement, and possibly refusal of conventional study.

Treating all gaps as methodological misreads the most interesting cases. A structural gap that you address with a standard study design is likely to reproduce the structural problem. The classification step protects you from that.

A worked synthesis example

Two studies of adolescent vaping uptake. Study A: an RCT showing peer-led messaging reduces uptake by 15% at 6 months in two suburban high schools. Study B: an ethnography of vape-shop culture in a rural community showing that for some adolescents, vaping is performative identity work tied to local in-group dynamics.

Appraisal: Study A by post-positivist criteria (well-powered, low attrition, internal validity strong; external validity to non-suburban populations limited). Study B by interpretivist criteria (rich thick description, credible interpretation, transferability cautious).

Synthesis claim: "Peer-led interventions show promise in suburban school contexts but are unlikely to scale uniformly across rural or identity-driven contexts. Interventions in identity-driven settings need to address the performative function vaping serves, not just the informational gap. Effective scale-up may require context-specific intervention design rather than uniform programs."

That claim is integrative, falsifiable, and policy-relevant. It is the synthesis.

When a review should take a position

Some reviewers worry that "taking a position" sacrifices objectivity. The honest reply: every synthesis takes a position; the question is whether the position is explicit or hidden. A review that papers over divergence implicitly takes the position that the divergence isn't important. Making the position explicit is more rigorous, not less.

The aim is defensible position-taking — claims supported by transparent appraisal and clear reasoning, open to challenge by reviewers and readers.

Closing

Synthesis is claim-making; comparison is preliminary work. Appraise studies by their own paradigm's criteria. Reviews should make integrative claims that no single source could make alone. Classify gaps as methodological, paradigmatic, or structural — they call for different responses.

Next: designing for complexity — adaptive design, validity across paradigms, and methodological integration.