Identity After the Uniform: What the Evidence Tells Me, Where the Gaps Are, and What I'm Doing Next

Identity After the Uniform is my research programme into how people rebuild a sense of identity after a defining occupational role ends. I started with military service, and I'm extending the work to police, fire, ambulance, NHS, prison, elite sport, addiction recovery, homelessness and medical retirement. From this research I've developed AURIS, my Identity Performance Psychology framework. AURIS proposes that identity, alongside employment, housing, education and health, should be recognised as a legitimate outcome of successful transition.

This report brings together three things I've never set out side by side before: what the wider academic literature already establishes, what my own interviews are telling me, and where my real gaps are. I'm publishing all three together because I think the case for this work is strongest when I'm honest about what I don't know yet, not just what I do.

My headline finding is straightforward. The evidence that identity disruption is real, common and currently unsupported is strong and consistent across everyone I've spoken to. My ability to measure it, with a validated instrument, at scale, in a way that would satisfy outside scrutiny, doesn't exist yet. That gap, not any doubt that the problem is real, is what stands between where I am now and where I need to be to support funding, policy or clinical claims. Closing it is my next major piece of work.

The thinking behind AURIS didn't appear from nowhere. It sits inside a real, traceable line of academic work stretching back seventy years:

An established theory I haven't fully connected to yet

Identity Fusion Theory (Swann, Jetten, Gomez, Whitehouse & Bastian, 2012, Psychological Review) is a validated academic construct for exactly what I mean by “how central the uniformed role was to identity”: a visceral sense of oneness between personal and social identity. This isn't a claim I need to defend, it's established, peer-reviewed, replicated science, and I can cite it with full confidence. It already has multiple validated instruments (a 7-item verbal fusion scale, a pictorial scale, and a continuous computerised version), none of which I currently use. One of my own interviewees, Matthew Addison-Black, has doctoral research built directly on this theory, which gives me a genuine, citable academic bridge rather than a coincidental parallel. I can see both low-fusion and high-fusion patterns in my own interviews, but I've only ever inferred this qualitatively. I haven't yet measured it with the validated scale that already exists for exactly this purpose, and that's a clear next step for me.

Sector by sector, where the outside evidence stands

Worth flagging on that last row: the Centre for Evidence for the Armed Forces Community, funded by the Forces in Mind Trust, is running a study right now specifically into the benefits partners of ex-Service personnel experience, called the OUTCOMES partners study. It won't have results until June 2027 at the earliest, so I'm not treating it as evidence yet, but its own reference list pointed me to five further sources I've now checked and verified, including one qualitative study that names identity loss as a challenge partners themselves face, not just veterans. It's also a good sign: a well-funded, established research centre is asking a version of the same question I am, just from the partner's side rather than the veteran's.

One of the sources I checked, a major RAND Europe and QinetiQ study on general transition support, reports real, significant progress over the past decade in practical transition support, and most people now transitioning successfully. I want to be precise about what that does and doesn't mean for my own claims: it's about general, practical support, employment, housing, how the process is communicated, not identity support specifically. My own finding that zero institutional identity-specific support currently exists still stands. I'm noting this here so I never let the two get conflated, by myself or by anyone reading this.

Everything below is drawn from the interviews I've coded so far, out of the 108 I'm aiming for in total. This is emerging evidence from an active, ongoing programme, not a final or population-wide finding, and I say so wherever I use it.

The ten patterns I'm seeing across interviews

How identity actually seems to unfold

Reading the interviews together, rather than one at a time, shows me seven distinct identity trajectories. I only originally built AURIS around two of them:

• Loss / Disruption: my default pattern, well-evidenced, the baseline case across most interviews
• Reconstruction: identity actively rebuilt through the five pillars after disruption, the core pattern AURIS is designed to support
• Continuity: identity was never strongly fused to the role, so its ending doesn't disrupt; a real pattern, but I haven't yet found a clean military example of it
• Expansion: the new identity genuinely grows beyond the role-bound one it replaced, not just recovers; one plausible case so far, and I want to test it against more data
• Substitution: a new role fills the space of the old one without deeper identity work; one well-documented case, self-described as still “partly mission-fused”
• Liberation: the end of the role is experienced as relief, not loss; one clear case, and an important reminder that loss isn't the only story

Alongside my default Identity to Community to Purpose to Stability sequence, I'm also seeing at least two alternative orderings: identity work done before leaving that prevents disruption rather than just speeding recovery from it, and stability reached by replicating the old role's structure rather than through the fuller sequence. A third possible ordering, where community comes first and identity clarity follows, hasn't shown up in my data yet. I'm treating that as a genuinely open question, not a closed one.

What I'm not tracking yet, but should be

Six factors keep coming up as plausible reasons some people transition more smoothly than others: whether the exit was voluntary or forced, whether identity-specific preparation happened before leaving, how strongly someone's identity was fused to their role, whether a community or social anchor was available afterwards, financial security at the point of transition, and health status at transition. None of these is a structured, comparable field in my data yet. Turning them into one is one of the highest-value, lowest-cost improvements I can make.

AURIS is built on five pillars: Awareness, Understanding, Regulation, Identity and Stability. These describe how I believe identity gets rebuilt after role loss. Mapping each pillar against both outside theory and my own interview evidence gives me an honest picture of where the framework is strong and where it's still my own working hypothesis:

Here's the honest reading of that table. Identity is my strongest pillar only relative to four others that all still need their own outside grounding done. It's a genuine, real strength relative to the rest of my framework, but it isn't yet evidence that AURIS as a whole has cleared outside academic scrutiny. I think saying that plainly is part of doing this properly.

The measurement gap

This is the single most consequential gap in my work. I don't yet have a validated instrument for my own core construct:

• My working Identity measure is a short, unvalidated pre/post self-rating across the five pillars. I'm explicit that it isn't a validated clinical scale.
• My in-app Identity Adaptation Index has essentially no validation data behind it yet.
• This isn't simply an oversight on my part. My own White Paper Vol.4, “Where Identity Work Belongs,” argues that identity after role loss is genuinely unclaimed theoretical territory, which is part of why no existing off-the-shelf tool fits cleanly.
• Validated instruments already exist for adjacent constructs, the Meaning in Life Questionnaire for Purpose, the General Belongingness Scale for Belonging, Schwarzer and Jerusalem's scale for Self-Efficacy. I haven't adopted any of them yet. They remain candidates, not decisions.
• I still have open questions underneath this too: is Purpose a separate outcome from Identity and Stability, or part of them? Is my idea of “self-trust” the same as general self-efficacy, or something narrower and more identity-specific? I need to resolve both before I can choose an instrument with confidence.

The sector-coverage gap

I have zero ambulance-sector and zero NHS-clinical-sector interviews, despite naming both as core populations, alongside military, police and fire, in every one of my white papers and in my own research questions. I've already decided to actively recruit into both sectors, and I've drafted standard caveat wording for all new outreach in the meantime. Material I've already published isn't retroactively covered by that fix, and I'm not pretending otherwise.

Specific gaps still open

Four separate lines of evidence point the same way, and each one stands on its own:

The need is real, and it's currently unmet

Zero institutional transition support is the single strongest, most-repeated finding in my entire dataset. It recurs independently across military, police and medical-discharge populations, with no exception anywhere in my interviews so far. People tell me this themselves, and they draw the unfavourable comparison to formal resettlement provision without being prompted to. This isn't a framing choice on my part. It's the most consistent single pattern the data contains. Qualitative convergence across sectors, zero disconfirming cases, and independent corroboration in the external police and medical-discharge literature is about as strong as this kind of evidence gets. I state it plainly, not as an emerging pattern.

The theoretical space is genuinely open, not crowded

My own literature mapping shows a coherent seventy-year line running from Erikson and Frankl through social identity theory, role-exit theory and post-traumatic growth, into a distinct military-transition research field from the 2010s onward. But identity after role loss, as a named, trackable outcome alongside employment, housing, education and health, is still genuinely unclaimed territory. Identity Fusion Theory offers me a validated, existing bridge into this space that I haven't fully used yet.

No existing tool measures what needs measuring

This is the sharpest version of the case for doing this work, not against it. Because no validated instrument currently captures what I mean by identity rebuilt after role loss, the fact that I don't have a validated instrument of my own yet isn't evidence of an unfinished side-project. It's the actual research frontier I exist to occupy. My task now is building and validating that instrument, not searching harder for one that already exists. This is my most defensible claim in the whole report, because it isn't an interpretation, it's a literature-search finding: it can be demonstrated, not just argued, and anyone can check it by looking.

The translation gap is a distinct, corroborated problem

Separately from the identity question itself, my own interviews and the external police-transition literature both independently identify a translation problem: civilian employers don't recognise or correctly value military or police experience, and sometimes actively penalise it. This compounds identity disruption with a real economic harm, and gives me a second, distinct reason to act alongside the identity-support case. Two unconnected sources reaching the same finding independently of each other is meaningfully stronger evidence than either one alone.

Here's my roadmap, in priority order:

• Close the measurement gap first: I'll resolve my open definitional questions, then decide whether to adopt validated instruments for the adjacent constructs while I build and validate my own Identity instrument, since nothing off-the-shelf covers it
• Actively resource ambulance and NHS-clinical recruitment: this is a decided action for me now, not an open question
• Turn my six candidate moderators into structured fields: voluntary versus involuntary exit, pre-emptive identity work, fusion level, community anchor, financial security, health status. This is a low-cost, high-value improvement to my data quality going forward
• Verify my one outstanding statistic: before I use it in any external-facing material
• Start my workshop delivery log now: nearly two months of live delivery data is currently unrecorded, and the longer I leave it the harder it is to recover
• Route new theoretical work through independent review promptly: so evidence-complete work isn't held up only by an unscheduled step on my end

A note on how I write about my evidence

I describe my findings as emerging evidence, not proof, for as long as that's what they are. Where a statistic or claim hasn't been independently verified, I say so and I don't use it externally until it has been. I'd rather publish something modest and accurate than something impressive and wrong.

This report reflects my research as it stands in September 2026, and I'll update it as my dataset grows toward its full target.