Free and Frank Advice: Four Fault Lines Over Mandate, Challenge, Occasion and Timing

11/04/2026

I describe the shape of the six-posture solution and the four fault lines that run through it, over mandate, challenge, occasion and timing, and I explain why each one persists as a dilemma rather than a mistake one side is making.

This is the fourteenth of twenty posts drawn from my doctoral thesis on free and frank advice in Aotearoa. Having described what the six postures share and how each holds the obligation, I turn here to where they part company. This post covers the uneven shape of the solution and the four fault lines that run through it, and what the literature says about why each one stays open.

The six postures share a commitment to practical statecraft, but they part company over the conditions under which free and frank advice is supported or constrained, and in my thesis I call those recurring divisions fault lines rather than disagreements, because the same division returns across the arrays, in the same place, with the postures arranged along it in a stable order. I do three things in this post: first, I describe the shape of the solution the fault lines run through; second, I set out the two fault lines that concern mandate and challenge; and third, I set out the two that concern the occasion of free and frank advice and its timing, ending with what the four together say about the institutional conditions that turn free and frank advice on or off.

The shape the fault lines run through

The solution is uneven, and that unevenness is part of what it shows. Four postures form the main body of the field, with the Institutional Pragmatist defined by 27 sorts, the Democratic Constitutionalist by 12, the Moment Chooser by five and the Collaborative Sense-Maker by four, which together account for 48 of the 50 defining sorts, while the Political Servant and the Westminster Protector sit at the outer edge, each defined by a single sort and retained because it met the tests set for the solution rather than because it was well populated.

The eight sorts that defined no posture show that the edges between postures are open, and their composition is telling. Five of the eight were new professionals and three were experienced state servants or senior public figures, and in my reading the newer practitioners may recognise the established ways of holding the convention without yet having resolved into one of them, while some experienced practitioners appear to carry more than one at once. The postures, read this way, are structured orientations with traversable edges, and the fault lines are the places where the differences between them are sharpest.

Mandate and challenge

The first fault line runs through the ministerial mandate, and the statement that carries it, Statement 8, that officials’ free and frank advice must fit a minister’s democratic mandate, separates the solution more sharply than any other. The Westminster Protector places it at +4 and the Institutional Pragmatist at +3, while the Collaborative Sense-Maker and the Moment Chooser place it at −5 and the Democratic Constitutionalist at −3, and the Political Servant, which places the mandate statement itself at a mild +1, carries its stronger commitment through the neighbouring statement about helping ministers deliver their political vision, at +3. One side reads fit with the mandate as part of what makes advice useful in a democracy, and the other reads principled independence as the point of giving advice at all.

The stake in this first fault line is democratic consent, and the literature helps explain why the division persists. Under legal-rational authority, power runs through the office, and the authority an election confers passes down a chain of delegation from Parliament to the official (Weber, 1947), yet an election settles who governs and leaves open how much deference that authority can command once advice is being formed. Responsiveness, on Mulgan’s (2008) account, is proper up to a point, with its limits set by the official’s obligations to future governments and to standards of legality and honest administration, and in my reading the mandate fault line is that question worked out posture by posture, with each pole carrying a democratic claim the other cannot absorb.

The second fault line runs through challenge, carried by Statements 12, 15, 16 and 25, and the division concerns where challenge belongs more than whether officials should push back. The Democratic Constitutionalist places challenge at the agreement end and treats it as a duty owed to the office, the Political Servant and the Westminster Protector hold it in reserve, the Moment Chooser accepts it on condition of timing, and the Institutional Pragmatist and the Collaborative Sense-Maker hold the middle. The sort itself shows the division over challenge, and the post-sort explanations and validation showed that the deeper question was its form, whether challenge should be placed on the record or kept in protected counsel between minister and official.

That division follows closely the distinction between open and closed models of free and frank advice (Mulgan, 2012). In the open model, advice is produced at some distance from the minister and remains subject to scrutiny and disclosure, and in the closed model it is confidential counsel given within a relationship of trust, where disagreement is possible and necessary but depends on staying behind closed doors. Research on the Official Information Act found officials committing less sensitive material to paper when they doubted it would stay protected, with difficult advice moving into oral conversation (Kibblewhite & Boshier, 2018; N. White, 2007a), and in my view that finding shows each side of the fault line protecting a condition the other needs, the record through which the state remembers its own reasoning and the confidence within which its hardest reasoning is done.

Occasion and timing

The third fault line runs through the occasion of free and frank advice, whether it is the ordinary work of governing or a protective warning kept for serious risk, and Statement 28 produces the cleanest split in the solution. The Political Servant and the Westminster Protector place the ultimate-failsafe statement at +5, the Democratic Constitutionalist at −4, and the Institutional Pragmatist and the Moment Chooser at −3, with the Collaborative Sense-Maker at a mild +2, so the solution splits three arrays against three, with strong agreement confined to the two single-sort postures.

Republican accounts of non-domination help explain what is at stake in that split (Pettit, 1997). The routine position is alert to ministerial power silencing official judgement before it is spoken, as officials anticipate what will be tolerated, while the protective position is alert to the opposite danger, of unelected expertise turning ordinary disagreement into a standing claim to supervise political judgement. In my reading the two rest on different theories of prevention, one seeking to head off the conditions that make a final warning necessary, and the other keeping a final register in reserve for the day those conditions arrive, and the convention appears to need both.

The fourth fault line runs through timing, carried by Statements 17, 19, 20 and 21, and it concentrates around one posture. The Moment Chooser rises to +5 on the experienced adviser who selects the moment to speak, the Institutional Pragmatist holds the timing statements positively but lower, the Political Servant places knowing when to offer advice at +4, the Democratic Constitutionalist sits near the neutral point, and the Westminster Protector places the experienced adviser’s selected moment at −3. For the Moment Chooser timing is the craft that makes advice hearable, while for the more cautious postures the same craft can become a way of trimming the advice, sparing the minister or delaying candour until the decision has already formed.

This fault line is the best recognised of the four in validation and the least theorised in the literature, and for that reason the thesis closes it with tests rather than a settled account. The first test is whether the substance of the advice remains intact, the second whether the advice enters before the decision closes, the third whether an adequate record remains, and the fourth whether the adviser could explain the timing as a judgement directed toward the quality of the decision. Together they ask whether the choice of moment served the decision or the adviser, and in my view that is the right question for a discretion that is exercised largely out of view, inside the relationship and on the adviser’s own anticipation.

Read together, the four fault lines answer the third research question conditionally, and that conditional answer is, I think, the most practical finding in this part of the thesis. The same institutional condition turns free and frank advice on for one posture and off for another: conditions that strengthen ministerial direction tend to support the Westminster Protector, the Political Servant and the Institutional Pragmatist while constraining the Democratic Constitutionalist, the Collaborative Sense-Maker and the Moment Chooser, and conditions that reward routine frankness support the Institutional Pragmatist, the Democratic Constitutionalist and the Moment Chooser while constraining the two protective postures. A single list of the conditions that support free and frank advice would therefore mislead, because whether free and frank advice is given depends on how a condition meets the posture a practitioner brings to it.

In the next post I bring the postures, practical statecraft, the shape of the solution and the fault lines together as the bounded plurality finding.

References

Kibblewhite, A., & Boshier, P. (2018). Free and frank advice and the Official Information Act: Balancing competing principles of good government. Policy Quarterly, 14(2), 3–9.

Mulgan, R. (2008). How much responsiveness is too much or too little? Australian Journal of Public Administration, 67(3), 345–356.

Mulgan, R. (2012, May 30). What future for free and frank advice? [Address]. Institute of Public Administration New Zealand Centennial Lecture, Wellington, New Zealand.

Pettit, P. (1997). Republicanism: A theory of freedom and government. Oxford University Press.

Weber, M. (1947). The theory of social and economic organization (A. M. Henderson & T. Parsons, Trans.). Free Press. White, N. (2007a). Free and frank: Making the New Zealand Official Information Act 1982 work better. Institute of Policy Studies.