It’s Not Just About the Evidence: The Interpretative Hinge Unpacked

19/02/2026

For three months now, I’ve been building one argument: the gap between policy advice and implementation isn’t just institutional or technical, it’s also interpretive. The people designing policy, the people delivering it and those impacted by it, operate from different understandings of what the problem is, what the policy means, what the solution is, and therefore what they’re supposed to do. Today’s post unpacks what I mean by interpretive and the interpretive hinge.

The pattern repeats. As Elizabeth Eppel documented, the Land and Water Forum spent seven years building sophisticated consensus across multiple stakeholders: from farmers, environmentalists, scientists, to Iwi and Hapū, to industry groups. It produced 150+ recommendations. Then, fewer than half were adopted. Environmental groups withdrew. The Forum collapsed. Not because the advice was bad, but because the delivery systems had never been in the room. When the recommendations arrived, they made no sense given the interpretive frameworks that regional councils, compliance officers, and enforcement agencies were operating from.

It is the same story with the leaky homes crisis. Derek Gill documented how the performance-based building code looked modern on paper but had no feedback loops to detect how councils and builders were interpreting the new regime until homes started rotting.

Just before Christmas, Christian Gade showed us why this potentially keeps happening. The same evidence can support completely different policies depending on what you value. Child poverty statistics can justify expanding welfare or cutting it. The data doesn’t change. The interpretation, shaped by ideology, beliefs, values, and assumptions, can change everything.

But Gade didn’t explain why interpretation is inevitable. That’s what we need now. That’s what today’s post is about. Because once we see why interpretation is built into knowledge itself, we understand why rebuilding the interpretive hinge isn’t about better communication, more consultation, or another co-design workshop. It’s about institutional capacity to do interpretive work transparently and well.

Today, I’ll argue three things. First, that interpretation is inevitable because there are no “brute facts”: all evidence is seen through interpretive lenses shaped by beliefs, values, and traditions. Second, that this explains why the hinge between policy advice and implementation keeps breaking: different actors operate from fundamentally different interpretive frameworks, and we have very little institutional capacity to translate between them. And third, that rebuilding the hinge requires more than better communication. It demands explicit acknowledgment that both advice and delivery are interpretive processes that work as translation mechanisms that connect advisory sophistication to delivery reality, and feedback and learning loops need to run from the beginning rather than being tacked on at the end. By the end of this post, I hope to have made the case that the interpretive hinge isn’t academic theory but the missing infrastructure our practical state needs to function.

Why There Are No Brute Facts

The answer starts with something that sounds academic but is actually just realistic. There are no “brute facts”: no neutral, objective data that exists independently of human interpretation. Every piece of evidence is seen through somebody’s lens, and the lens shapes what you see.

Let’s be real: “evidence-based” has become a rhetorical weapon. Check your socials: it’s full of claims that policies are “not even close to evidence-based” and “so they are ignoring the evidence”. Usually from people who just disagree with the decision. Evidence is a much more elastic term than anyone wants to admit.

If you don’t believe me, think about freshwater policy. The same water quality data can tell completely different stories depending on who’s interpreting it. Farmers might see acceptable nutrient levels given current land use. Environmental scientists might see degradation trends that require urgent intervention. Iwi might see impacts on mauri that the scientific metrics don’t capture at all. Recreational users might see declining swimming quality. Industry might see regulatory costs that threaten viability.

The data—nitrogen levels, E. coli counts, water clarity measurements—appears objective. But what counts as “acceptable”? What threshold requires action? What trade-offs are reasonable? Every one of these questions requires interpretation shaped by what one values, what knowledge tradition you’re operating from, and what you believe about how freshwater systems work.

This is exactly what the Land and Water Forum was designed to address. Nearly 60 stakeholders with fundamentally different interpretive frameworks spent seven years in structured dialogue, building shared understanding across those differences. As Elizabeth Eppel documented, this wasn’t usual consultation theatre. It was systematic integration of diverse evidence and expertise into coherent policy advice. They had to face each other directly, as Alistair Bisley put it, and “take the responsibility of reconciling the various things that they want.”

And it worked, at the advisory level. They built consensus across interpretive frameworks that had been in conflict for decades. That’s genuinely sophisticated policy advisory work.

But then it hit the wall of implementation. Fewer than half the 153 recommendations were adopted. Environmental stakeholders withdrew in 2016. The whole thing collapsed.

The Advisory System is Already Contested

Why? Because while the Land and Water Forum successfully navigated contested interpretations at the advisory level, it never connected to the delivery system where those same contests would play out again.

Jon Craft and Brian Head’s work helps us understand what happened. Policy advisory systems aren’t just plural: they’re contested. Different sources don’t just provide different information. They operate from fundamentally different interpretive frameworks about what counts as evidence, what problems matter, and what solutions make sense.

The Land and Water Forum proved that sophisticated advisory processes can manage this contestability. Federated Farmers and Forest and Bird sat in the same room for seven years and found common ground. That’s remarkable. But managing contestability in advice is only half the job. You also have to manage it in delivery.

Regional councils weren’t operating from the consensus framework the Forum built. They had their own interpretive traditions: shaped by matauranga, resource management law, local politics, some technical capacity, and relationships with local communities. Compliance officers had their frameworks: shaped by regulatory and enforcement experience, legal constraints, and practical limitations. Community groups had theirs: shaped by local water quality experiences and trust or distrust in official processes.

None of these delivery actors had been part of building the Forum’s consensus. When sophisticated recommendations arrived, they had to interpret them through their existing frameworks. And those frameworks often made the recommendations look unworkable, threatening, or simply irrelevant.

The hinge between sophisticated advice and messy implementation simply didn’t exist.

Why Action Follows Belief

Here’s the core issue. You cannot understand what people do without understanding what they think they’re doing.

Action follows belief.

Compare two cases. The Land and Water Forum failed despite advisory sophistication because delivery systems weren’t part of the interpretive work. The Māori Communities COVID-19 Fund succeeded despite being designed at speed because it explicitly worked with different interpretive frameworks rather than trying to impose one.

The MCCF didn’t tell Iwi and Māori providers what to do. It trusted them to interpret what their communities needed. Some focused on food security. Others on digital connectivity. Others on mental health support. Still others on keeping tamariki connected to school. The MMCF “policy” wasn’t a set of instructions: it was an authorising framework that allowed different actors to act on their beliefs about what good looks like in their specific contexts, while having a eye firmly on a specific outcome. Even then, our evaluation of the MCCF found, the dominant narrative wasn’t about hitting targets, though the investment exceeded its targets, it was about manaakitanga: getting a “foot-in-the-door of some whare” and rebuilding trust and confidence where the mainstream systems had failed, as a pre-cusor to a korero about vaccination.

The MCCF worked because it acknowledged from the start that different communities and different whanau in those communities, would interpret “vaccination support” differently, and that this interpretive diversity was a strength, not a problem.

The Land and Water Forum tried to use it’s interpretive diversity through consensus-building. That worked brilliantly for producing policy advice. But implementation requires ongoing interpretation. Rules get applied to specific situations. Trade-offs get made in local contexts. Competing values get balanced in real decisions about real farms and real rivers. You can’t consensus that away in advance. The interpretive and reinterpretive work has to keep happening in delivery.

Why Traditions Shape Everything

These beliefs and interpretations don’t emerge from nowhere. They’re inherited from traditions: professional training, institutional experience, cultural background, local context and how communities understand the world around them.

When farmers look at freshwater policy, they’re drawing on traditions about land stewardship, economic viability, and family farm succession. When environmental scientists look at the same policy, they’re drawing on traditions about ecosystem health, scientific evidence, and precautionary principles. When Iwi me Hapu look at freshwater policy, they’re drawing on traditions about whakapapa, kaitiakitanga, and relationships with water that predate the entire debate.

The Land and Water Forum’s achievement was creating structured space where these different traditions could encounter each other, understand each other, and find areas of common ground. That’s valuable work. But traditions don’t disappear once consensus is reached. Regional councils have their own traditions: shaped by decades of resource management practice, relationships with local communities, and constraints of local government funding. Those traditions didn’t change just because stakeholders reached agreement in Wellington.

We’ve seen this pattern repeat. The Living Standards Framework assumed all agencies would adopt Treasury’s framework for understanding wellbeing. But frontline agencies operated from their own professional traditions about what constitutes good. The leaky homes crisis happened partly because councils operated from traditions of minimum compliance checking whilst the new regime assumed they’d actively assess innovative building methods. In each case, the policy assumed a uniformity of interpretive tradition that simply didn’t exist in delivery.

This is where Aotearoa gets particularly complex. We don’t just have professional traditions. We have constitutional ones. The Land and Water Forum operated within a Crown policy process with Westminster sovereignty, ministerial authority, and public service advice. But Te Tiriti establishes something different: the need to constantly balance rangatiratanga and kawanatanga. So when Iwi participants engaged in the Forum, the question became were they operating as stakeholders in a Crown process or as Treaty partners with their own authority? That ambiguity was never resolved. And, maybe it cannot be. But suffice it to say, when the delivery systems tried to implement the Forum recommendations, that same ambiguity played out at regional level, creating confusion about which Iwi me Hapu held what authority and on what basis. The MCCF lesson suggests a different approach is possible: one that acknowledges from the start that Te Tiriti creates two different sources of legitimacy that must be negotiated, not collapsed into one. And, not to belabour the point, but the MCCF was successfully implemented.

What This Means for the Hinge

So what might rebuilding the interpretive hinge actually require? Drawing on what we’ve seen across these cases, I want to propose four elements. These are my initial ideas only.

First, we might need to acknowledge explicitly that all advice and all implementation decisions are interpretive. The Land and Water Forum did this well in the advisory space: it recognised that different actors had different frameworks and created processes to negotiate them. But it never extended that recognition to delivery. Implementation was treated as technical execution, not as ongoing interpretive work. The leaky homes crisis suggests what happens when we don’t acknowledge this: no one realised that councils and builders were interpreting the new regime in fundamentally different ways until homes started rotting.

Second, we might need processes that translate between interpretive frameworks, and sustain them through implementation, not just policy development. The Forum created brilliant translation processes among stakeholders. But once recommendations were made, translation stopped. Delivery actors had to figure out what the consensus meant in their contexts, with no Forum process to help. This seems to be what Cairney was getting at when he argued perfect coherence is impossible: maybe the real work is building capacity for ongoing translation rather than trying to achieve alignment once and for all.

Third, we might need to maintain standards while acknowledging interpretation is inevitable. Some freshwater outcomes are objectively better than others. Some policy approaches fit evidence better. But recognising this doesn’t eliminate interpretation, it just means we need transparent criteria for judging between competing interpretations. The Forum built these for advice. What delivery seems to need is similar criteria: ways to assess whether different regional and local interpretations of national policy are legitimate adaptations or problematic deviations. The Living Standards Framework attempted this with its domains and indicators, but couldn’t bridge the gap to operational meaning.

Fourth, we need feedback and learning loops from delivery to advice that operate from the beginning, not as an afterthought. This is where the Land and Water Forum failed most fundamentally, and it’s a pattern we’ve seen elsewhere. Seven years of sophisticated consensus-building, and delivery systems only learned about it when recommendations arrived. The leaky homes crisis had the same structure: policy changes happened at the centre without feedback loops to detect how they were being interpreted on the ground until it was too late. By contrast, the MCCF built feedback into its design: regional Te Puni Kōkiri staff acted as boundary spanners or what I call as “weavers”, constantly translating between Crown authorising and community delivery.

I’m proposing that what connects these cases is the presence or absence of what I’m calling the interpretive hinge: the institutional capacity to recognise that policy and delivery are both interpretive processes, to translate between different interpretive frameworks, to maintain standards without claiming neutrality, and to create ongoing dialogue rather than one-way transmission. That’s a hypothesis based on the pattern I’m seeing across these cases, not a definitive conclusion.

The Infrastructure We Need

If this working hypothesis holds, then the interpretive hinge isn’t a nice-to-have. It’s core infrastructure the practical state needs to function. The Land and Water Forum, the Living Standards Framework, the leaky homes crisis, and the MCCF all suggest we can build sophisticated advisory processes. What they also suggest is that advisory sophistication alone isn’t enough. You have to connect advice to delivery from the beginning. You have to recognise that implementation is interpretive work, not just technical execution. And you have to create institutional capacity to sustain translation and negotiation all the way through.

Without that capacity, we seem to keep building elegant policy advisory processes that produce sophisticated recommendations that collapse on contact with messy reality. The question is whether we can build institutions that do interpretive work well, transparently, and in ways that connect advice to action. That’s what the rest of this series will explore.

Next month, we’ll see how decolonial and feminist approaches use these interpretive methods to expose whose knowledge counts, whose stories are heard, and whose power is protected by claims to neutrality.

References

Bevir, M., & Rhodes, R. A. W. (2003). Interpreting British governance. Routledge.

Eppel, E. (2011). Illustrative case: Land and Water (Protection and Use) Forum. Future State 2: Experimentation and Learning in Policy Implementation – Implications for Public Management.

Eppel, E. (2013). Collaborative governance case studies: The Land and Water Forum. Working Paper No: 13/05. Wellington: Institute for Governance and Policy Studies, Victoria University of Wellington.

Eppel, E. (2014). Improving New Zealand water governance: Challenges and recommendations. Policy Quarterly, 10(3), 66-73.Gade, C. (2024). When is it justified to claim that a practice or policy is evidence-based? Reflections on evidence and preferences. Evidence & Policy, 20(2), 244–253.

Gill, D. (2016). Regulatory coherence: The case of New Zealand. In D. Gill & P. Intal Jr. (Eds.), The development of regulatory management systems in East Asia: Country studies (pp. 175–227). Jakarta: ERIA (Economic Research Institute for ASEAN and East Asia).

Head, B. W. (2023). Reconsidering expertise for public policymaking: The challenges of contestability. Australian Journal of Public Administration, 83, 156–172.

Footnote:

A caveat on “evidence-based policy.” I’m not arguing evidence doesn’t matter: of course it does. What I’m challenging is the rhetorical and political power the word “evidence” has accumulated. It is often used to determine which research projects get funded and which don’t, whilst systematically ignoring the insights of those who deliver services, produce regulatory outcomes, and hold practical knowledge and local context. As Brian Head (2023) and others have demonstrated, expertise in policy advisory systems is contested and distributed: it lives in communities, with regulated parties, with Iwi and Hapū, with frontline workers, not just in university research departments. Our problems are complex, inexplicable, interdependent, and some are intractable. Atua help us if we have to wait for scientists to tell us what we’re already seeing on the ground. The interpretive work I’m describing isn’t a replacement for rigorous evidence: it’s the capacity to recognise that evidence always requires interpretation, and that the people doing implementation often hold knowledge that’s as important as what sits in academic science journals.