The Free and Frank Series: The Ghost in the Machine

11/10/2025

This is the second instalment in a series exploring what free and frank advice means in Aotearoa’s institutional arrangements. Last week, I introduced the research journey. This week, I turn to the puzzle itself: how something so central to our democratic machinery became so undefined, and why that indefiniteness appears to be by design rather than oversight. By the end of this post, I hope to have shown that free and frank advice operates as a “ghost in the machine,” omnipresent yet invisible, regulated yet undefined, and that its productive ambiguity may be what allows competing values to coexist within the same institutional framework.

Today, I begin with the mystery itself: how something so central to democratic governance became so deliberately undefined. From that starting point, I trace the productive ambiguity surrounding the concept of free and frank advice and introduce a metaphor that will carry the rest of the series: the ghost in the machine. The post closes by mapping the journey ahead, including the two broad movements through which the series will proceed.

Four years ago, I thought I was asking a simple question: what is free and frank advice, and why does it matter? It appears in our laws, gets mentioned in Cabinet manuals, and gets invoked daily across government as the gold standard for how officials should advise ministers. Surely someone, somewhere, had defined it properly. I was wrong. Spectacularly, fascinatingly wrong. What I discovered was far more interesting: a concept so central to how our democracy works that it appears everywhere, yet so deliberately undefined that it means completely different things to different people. It is not randomly different, but rather systematically and coherently different. The net result is effectively six distinct tribes speaking the same language but meaning entirely different things. Over the next six months, this series will offer accessible, plain-speaking commentary on key insights from my doctoral research into what constitutes free and frank advice. It won’t be the full academic treatment: think of the series as opening a window into the thinking and findings, not the finished scholarly work itself. This remains a work in progress, and I reserve the right to be wrong about some things.

Let’s get into it!

The Mystery

Free and frank advice occupies a unique position in Aotearoa’s constitutional and institutional arrangements. Unlike other Westminster jurisdictions, we’ve never had a major constitutional crisis that forced us to define precisely what we mean by the relationship between ministers and officials. We’ve improvised our way through constitutional development, creating conventions and understandings that work in practice but resist precise definition.

This improvisation has served us well. Our constitutional flexibility has allowed us to adapt to changing circumstances without the rigidity that formal constitutional documents might impose. However, it has also created spaces of productive ambiguity where important concepts remain deliberately undefined, and some would argue that it gives the Executive far too much power without corresponding accountability.

Free and frank advice is perhaps the most important of these undefined concepts. However, throughout my research, political neutrality came a close second. Followed, perhaps unsurprisingly for these times, by merit-based appointments and collective responsibility. Free and frank advice, however, seems important; for it sits at the heart of the relationship between political authority and administrative expertise, between democratic accountability and professional independence, between serving the government of the day and serving a future government and the broader public interest.

The concept emerged from our Westminster inheritance, but it has evolved in distinctively Aotearoa New Zealand ways. Te Tiriti o Waitangi and the ongoing negotiation between kāwanatanga and rangatiratanga shaped our constitutional development. This created a constitutional space where advice was never simply about neutral expertise serving political authority, but also about navigating complex relationships of power and obligation.

The historical origins reveal something fascinating about the concept of free and frank advice. It originally crystallised around protecting ministerial collective responsibility, effectively creating a “black box” where ministers could debate and disagree in private before presenting unified public positions. However, officials, being bureaucrats, eventually extended that same protection to themselves. What began as a shield for political deliberation to agree on collective responsibility became a shield for administrative deliberation, particularly for all manner of policy advice.

This expansion now faces new challenges. In our current environment of wicked problems and expanded policy advisory systems, with think tanks, consultants, communities, lobbyists, and various non-state actors all contributing to policy development, a crucial question is emerging. Do these new players also have the right to offer “free and frank advice” and enjoy the related protections? The ghost that once haunted just the relationship between ministers and officials is now spreading through an entire ecosystem of advisory relationships.

The Deliberate Ambiguity

The best way into that answer is a crucial insight, and that is that the ambiguity surrounding free and frank advice isn’t accidental. Rather, it’s by institutional design, whether conscious or unconscious. The ambiguity works precisely because it can accommodate different interpretations while maintaining enough shared meaning to enable coordination and accountability.

Consider what would happen if we defined free and frank advice precisely. A narrow definition emphasising political responsiveness or evidence might satisfy ministers. Still, it would concern those who worry about the long-term capacity of government institutions, and the often tedious work that keeps the lights on, but is not politically interesting. Similarly, a broad definition emphasising professional independence and evidence might satisfy officials, but would concern those who worry about democratic accountability, political control, and recognise that evidence itself is a social construction.

The current ambiguity allows both concerns to coexist within the same institutional framework. Ministers can interpret free and frank advice as responsive professional service that helps them achieve their political objectives. Officials can interpret it as independent professional judgment that serves broader constitutional values. Both interpretations can operate simultaneously without necessarily conflicting.

This productive ambiguity reflects what the literature identifies as a “boundary concept”: an idea that helps manage the tension between competing institutional logics without definitively resolving it. The concept of free and frank advice helps manage the tension between political and administrative authority without requiring a final resolution that might prove inadequate in different circumstances.

But that’s not enough. Not enough for accountability, not enough to scrutinise it, lift its quality, or get more or less of it. It’s not enough if it needs to be regulated. The very ambiguity that makes the concept work also makes it nearly impossible to improve, control, or even understand properly.

The Ghost in the Machine

As I have progressed the research, I’ve come to think of free and frank advice as the “ghost in the machine” of Aotearoa’s democracy. Something that haunts our constitutional arrangements: omnipresent yet invisible, essential yet elusive, regulated but undefined. Like all good ghosts, it shapes behaviour and outcomes while remaining stubbornly resistant to direct observation or capture.

This spectral quality isn’t a malfunction of our democratic machinery. It’s how the machine actually works. The ambiguity and contestation surrounding free and frank advice aren’t problems to be solved but potentially features of how democratic governance maintains flexibility and adaptability while preserving essential values.

The ghost enables our democracy to promise different things to different people while maintaining enough coherence to facilitate coordination and accountability. It allows competing values of ministerial responsiveness and independence, effectiveness and integrity, and political sensitivity and professional standards to coexist within the same institutional framework.

Why This Matters

Understanding this constitutional mystery matters beyond academic curiosity. How ministers and officials relate to each other shapes every government decision that affects our lives. Understanding these relationships better might help us improve how democracy functions, or at least understand why it sometimes fails to do so.

The research I’ll share over the coming weeks reveals that what appears to be confusion and disagreement about free and frank advice actually has underlying structure and logic. Using a multifactor analysis, I’ve been able to distil six distinct ways free and frank advice is understood, each representing a coherent philosophy of how policy advisory relationships should operate, in that moment when a difficult conversation needs to be had.

These aren’t just academic distinctions. They reflect genuine differences in how practitioners perceive their roles, responsibilities, and relationships. Understanding these differences can help explain why debates about advisory relationships often involve people talking past each other, and why reforms that seem obviously beneficial to some are obviously problematic to others.

What has emerged, then, is the outline of a mystery. Free and frank advice sits at the heart of democratic governance in Aotearoa, yet it has never been defined. Its ambiguity is not accidental but productive, enabling competing values to coexist within the same institutional framework. The ghost in the machine is not a malfunction; it is how the machine works. But to understand how it came to work this way, we need to trace the invisible contracts that govern the relationship between ministers and officials. Next week, we turn to the public service bargain.

The Journey Ahead

Across this series, we’ll explore this constitutional mystery from multiple angles. We’ll examine the theoretical frameworks that public administration scholars and political theorists use to understand advisory relationships, explore the methodological challenges of studying something that’s deliberately undefined, and dive deep into what practitioners actually think about free and frank advice.

The first part of the journey falls into two broad movements. The first, spanning roughly the next ten weeks, works through the public administration literature: the public service bargain, the purple zone, the authorising environment, and the political economy that furnishes them all. These are the frameworks that describe where free and frank advice operates and under what conditions. It is very much focused on the public administration literature.

The second movement turns to political theory: consent, rationalism, civic republicanism, and trust. These are the enduring theoretical traditions that ask big questions about the authority unelected officials have to challenge elected ministers, and whether that question can ever be satisfactorily answered. Together, the two movements build toward a methodological turn: how one might study a concept that resists definition by design.

As we progress, we’ll discover that the disagreement itself might be the answer. The contested nature of free and frank advice may precisely enable it to serve as a boundary concept that manages fundamental tensions in democratic governance.

We’ll see that this isn’t unique to Aotearoa or to advisory relationships in general. I offer an early hypothesis that democratic governance operates through an ecosystem of constitutional ghosts, which are deliberately undefined concepts that enable democratic institutions to function precisely because they remain ambiguous.

Most importantly, we’ll see that understanding how democracy actually works requires paying attention not just to formal institutions and explicit rules, but to the informal understandings and tacit agreements that make those institutions function in practice.

A Note on Method

The insights I’ll share draw from interviews with over 210 practitioners (elected and unelected officials from across the last nine administrations), conducted using qualitative interviews, Q-methodology, and validation wananga, workshops, and interviews. Q is a research approach specifically designed to study contested and subjective concepts. It isn’t speculation or theory; it’s grounded empirical research on how people actually think about advisory relationships and, in particular, free and frank advice. Across 2025, it involved me in analysing correlations between subjects’ Q-sorts, forming a correlation matrix. Then I performed an eigen-decomposition of the matrix to extract factors, stories, or perspectives representing shared viewpoints. Factor loadings indicate the degree to which each subject aligns with these factors, thereby explaining the variance. What I can promise with this method is intellectual honesty and precision about what linear algebra reveals, even when it’s inconvenient or surprising.

Next Week

Next week, we’ll begin mapping the theoretical landscape that surrounds advisory relationships. We’ll start with the concepts of Faustian and public-sector bargains to consider the implicit deals that shape relationships between ministers and officials, and to see how these bargains have evolved. The ghost in the machine isn’t going anywhere. It’s been haunting our constitutional arrangements for over a century, and it’s likely to continue doing so for the foreseeable future. Understanding how it works may be the key to understanding how democracy itself functions despite, or perhaps because of, its fundamental contradictions and tensions.