My submission is attached as a pdf file.
I write in my capacity as a 2017 recipient of the Sir Roland Wilson scholarship, which is facilitated
through the Australian Public Service (APS) and the Australian National University (ANU). Scholarship
candidates are selected based on the merit of proposals that will have direct and enduring relevance
to the APS. As such, my research, which is ongoing, examines the language of APS policy advice and
asks how it comes to be rebuffed by its ministerial audiences. I draw on a variety of Australian and
international case studies, in which I treat language as a symptom of institutional constraints and
requirements, such as the Public Service Act and Freedom of Information (FOI) legislation, as well as
the broader edifice of objectivity. In this vein, I would like to discuss the quality of written policy
advice, calls for changing FOI legislation, and international comparisons.
Recent reviews of the APS, such as Terry Moran’s Ahead of the Game (2010) and Peter Shergold’s
Learning from Failure (2015), were at pains to emphasise the importance of high-quality policy
advice and, with it, the ability to mount effective, candid arguments. Although the 2010 review does
not make specific recommendations related to this, by 2015, there is a suggestion that the FOI Act
be amended in order to prevent feeble presentation of written advice.
One of my case studies is the language of Australian intelligence assessments in the lead-up to the
Iraq war in 2003. Two agencies, the Office of National Assessments and the Defence Intelligence
Organisation, are examined. Neither agency is bound by FOI legislation. Indeed, each can feel
reasonably secure in only very rarely having to present their advice publicly. Although intelligence
and policy advice are not the same, the formulation of intelligence advice adheres to similar
expectations around being independent, forthright and robust. So what kind of language do we see
from agencies that do not need to worry about being FOI’d? On the one hand, DIO maintained its
position on the inconclusive nature of intelligence about Iraq’s WMD right up until the beginning of
the war in March 2003. On the other, ONA slowly moved from sharing a similar view around the lack
of clear evidence to becoming quite certain that Iraq had WMD, which aligned it more closely with
the government’s direction. Nonetheless, both were largely ignored by the then Prime Minister, who
preferred to source most of his public references to intelligence from Australia’s allies. Contrasting
the two agencies – ONA with a keenly responsive eye to the policy environment, DIO maintaining a
more objective distance from it – with the knowledge that neither is bound by FOI legislation,
suggests that arguments about its constraining effects are not persuasive. That is, neither the
unambiguous language of ONA nor the more nuanced nature of DIO’s advice had much of an impact
on how the government made its case for war with Iraq.
Arguments for amending FOI legislation suggest that, were it not for impediments like FOI, public
servants’ appetite for risk and their ability to argue frankly and fearlessly would be substantially
enhanced. In other words, the APS is already good at written advice but is constrained by forces
beyond its control. Yet we can see from the Iraq example that neither ONA’s increasingly robust
language nor DIO’s more impartial counterpart were effective. This leads one to wonder whether
objections to FOI are a way to rationalise the often nebulous nature of advice. Earlier in this
submission, I referred to the edifice of objectivity, and the frequent references made to it via
legislation, the APS Code of Conduct and various reviews. I argue that this requirement for
objectivity, although noble and well-intentioned, has created an ethos that exerts an inhibiting
influence on how policy advisers articulate and argue, resulting in a language that may well be too
weak to compete with louder voices, particularly amid a sea of ‘authenticity’ and opinion. The
requirement to be apolitical and objective seems to prevent advisers from realising that their
language does not have to be drained of life to be impartial. Facts do not speak for themselves; they
require context and argument. Yet, according to public policy academic Giandomenico Majone, “as a
firm believer in the virtues of the scientific method”, the policy adviser generally has “a distaste for
advocacy and persuasion”1. This is a mistake, in Majone’s estimation, because ignoring rhetoric and
advocacy (that is, unambiguously arguing for appropriate policy options) is likely to lead to failure:
[i]n policy analysis, as in science and in everyday reasoning, few arguments are purely
rational or purely persuasive. A careful blend of reason and persuasion is usually more
effective than exclusive reliance on one or the other2.
The perhaps self-imposed constraints created by focussing so resolutely on rationality, evidence and
objectivity have been observed not only in scholarship on public policy. Even those who might be
assumed to carry a more positivist mantle – scholars from disciplines like the sciences, statistics, and
economics, for example – have argued that what we have come to regard as objectivity is, perhaps
paradoxically, incompatible with reasoning that wants to persuade. Deirdre McCloskey, for instance,
has asked
[i]f some good economists espouse positivism the question arises how economics would be
different without it. That remains to be seen, though one thing is clear: economics without
positivism would be more, not less, rigorous and scientific, because it would have to face up
to more arguments3.
With its objective coating, the language of policy advice seems to pose as a window to a neutral,
clear-cut reality. Yet, as McCloskey insinuates, becoming more critically aware of how language is
constructed – such as its tendency to offer information as facts rather than educated interpretations
– could lead to stronger, more robust policy advice. Further, a language of objectivity can suggest
evidence where there is none or very little. For example, when objectivity is brandished by way of
evidence gathered through engagement with experts, this language may mask the fact that the
experts with whom one has consulted are not necessarily always the best, most knowledgeable
authority. At best, the result is incomplete or even faulty information camouflaged as objective fact;
at worst, policy is implemented based on information that is shallow and even unassessed by peers.
I can offer no easy fixes here and am not advocating for removing the requirement for objective
advice. For producers and users of knowledge, objectivity is an ideal that signals trust and legitimacy.
Without it, to quote Wildavsky, there is no “agreement on a starting place”4. I do, however, suggest
that it has become imperative for the APS to dig more deeply into what is preventing policy advice
from being more persuasive, and to understand that amending the FOI Act, for example, or sending
advisers to training that teaches them to use plain English, are not enough. As Greenhalgh and
Russell observe, the way to deal with the “inherent and irreducible messiness” of policy-making is
1
Majone, Giandomenico, Evidence, Argument, and Persuasion in the Policy Process, Yale University Press,
1989, Michigan, p. 36.
2
Ibid, p. 37.
3
McCloskey, Deirdre, Knowledge and persuasion in economics, Cambridge University Press, 1994, Cambridge,
p. 23.
4
Wildavsky, Aaron, Speaking Truth to Power: The Art and Craft of Policy Analysis, Little, Brown and Company,
1979, USA, p. 7.
not to produce more rigorous, more relevant, less ambiguous, more timely or more
appealingly presented evidence but for policy makers to develop a better awareness of their
own behaviour as players in the argumentation game.5
One possible solution to building such awareness, they propose, is for policy players to
“acknowledge and take account of their respective points of departure” and to make explicit “the
values and premises on which each side has built its case”. Another is to promote “the training of
policy making teams in the art of rhetoric”6. While this is usually anathema to civil servants who
assume rhetoric to be the opposite of objectivity, it simply means becoming more proficient at
logical argumentation. I suggest there is ample room for APS policy advice to get better at this key
ingredient in persuasive written advice.
Regarding the review’s proposed examination of international practice, I expect the panel will
compare Australia with other Westminster jurisdictions, which will undoubtedly yield some useful
insights. The UK Institute for Government and Behavioural Insights Team’s recently-published report
‘Behavioural Government’, for example, is an important exploration of officials’ predispositions
when shaping government policies and services. Further, New Zealand’s Policy Project, which places
policy advice at its core, and New Zealand’s Institute of Economic Research, which ranks policy
advice and makes recommendations for improvement, are both evidence of a government that
understands the fundamental connection between transparency and trust. We could do a lot worse
than to try similar approaches. However, while references to other Westminster jurisdictions will
likely be unavoidable, I suggest casting the net wider. For example, why not examine jurisdictions
that have managed to implement major reform of a kind unlikely to be passed here? In 2005, for
instance, the Netherlands moved away from incarceration due to prison overcrowding. Instead, the
country began to focus on rehabilitation and decarceration. You will no doubt have read that it
began renting out prison cells to inmates from neighbouring countries. How was this possible? What
kind of expert advice, collaboration and citizen engagement were required to achieve this? Similarly,
Estonia is now almost entirely digital. If its government had to move into exile, it could continue to
govern online. What did this small, former Soviet nation do to achieve such an intensely ambitious
reform? Surely Australia, as a developed, middle power, has the wherewithal to achieve innovation
of similar scope. It seems unlikely while we maintain the passivity of Westminster advisers.
This final point leads me to reflect on the APS of the future. Ironically, it relates to FOI, which I earlier
insinuated was not a barrier to good policy advice. I stand by this suggestion, but acknowledge that
it poses an interesting dilemma, which relates to APS reputation and, with it, public trust. Much
scholarship here in Australia has considered the two-fold effects of greater transparency in terms of
the APS’ longer-term credibility. The ANU’s Richard Mulgan, for instance, suggests that transparency
increases the likelihood that government deceptions will be uncovered and thus acts as a
spur to government honesty. On the other hand, the fact that more public service advice
may end up in the public arena places officials under greater pressure to compromise with
the truth in the interests of not undermining the credibility of their political masters.7
The former is a democratic mainstay. Yet, over time, he notes, the latter will undermine APS
expertise and reputation, citing institutional independence as way to protect these crucial qualities.
5
Greenhalgh, Trish and Jill Russell, ‘Reframing Evidence Synthesis As Rhetorical Action in the Policy Making
Drama’, Healthcare Policy, Vol. 1 No. 2, 2006, p. 40.
6
Ibid, p. 41.
7
Mulgan, Richard, ‘Truth in Government and the Politicization of Public Service Advice’, Public Administration,
Vol. 85, No. 3, 2007, p. 585.
He mentions the Australian Institute of Health and Welfare, the Australian Institute of Criminology
and the Australian Bureau of Agricultural and Resource Economics as examples of statutorily
independent government offices (we might add the Parliamentary Budget Office, the Productivity
Commission and the Commonwealth Grants Commission). My own research echoes these
sentiments, and I urge the panel to consider ways in which to invigorate and operationalise greater
commitment to APS independence. Without it, its reputation for integrity and accuracy – as well as
the public trust it may inspire – is vulnerable to corrosion.
Many thanks for the opportunity to contribute and I would be very happy to elaborate on any of the
points made here or in my research.
With best wishes,
Dr Christiane Gerblinger