Can philosophy help to assess and create public policy? The biggest push in recent years to suggest that maybe it can has been the growth of the idea of “evidence-based politics and policy”. This idea comes explicitly from methodological and epistemological reflection.
Unfortunately, it comes also from the tradition of positivism. “Evidence-based politics” can too easily be, therefore, an opponent of serious reflection; an opponent of profound cultural critique. It can tend to support through methodological fiat an essentially conservative “reformist” outlook: How, for instance, could you obtain evidence to validate a proposition such as “Our society needs a revolution right now: only a revolution could actually save us, now”? You couldn’t. Thus revolutionary change doesn’t get a look-in.
To be clear, there is an obvious and important role for evidence in the assessment of policy-interventions. But we believe that:
a. Our precise conception of evidence and the methods for gathering it needs to be sensitive to the nature of that which we are investigating. For example, establishing good evidence to support the claim that a population is happier under a certain political and economic system is very different to establishing good evidence for the efficacy of a biomedical intervention for a bacterial infection.
b. Evidence is only one part of the picture. Other factors such as precaution, political economy, asymmetry, and framing are also crucial.
Stepping back and thinking more philosophically and more deeply about politics helps one to build up a much more complete picture.
So here we hope to move beyond mere evidence base and set out what we believe to be the five parameters by means of which any public policy position or intervention ought to be assessed.
1. PRECAUTION. When someone steps out into the snow and ice, they are wise to exercise caution. It is slippery under foot, and a fall while alone in the cold might be ultimately fatal. That caution might be observed in the care a person exercises when taking each step, and in their tendency to hold onto things more than they otherwise might, and so on. To take such caution is wise.
But would we be willing to say of that person that they are wise if they didn’t take reasonable precautions too?
Perhaps they might take precautions: they might procure some ice grips to put over their shoes, or simply a pair of boots more suited to walking in ice and snow. They might purchase some Nordic walking poles. They’ll certainly wear warmer clothes in colder weather. They might even take the precaution of not taking unnecessary walks in icy conditions; they might for instance organise a stash of food to enable them to see out a cold snap. All reasonable precautions, one might say. To take such precautions is wise.
So, what might we say about politics and precaution?
The context is a little different but not such that the logic of caution and precaution does not apply. Here the caution and precaution differ in that in our example above the conditions that motivated our caution and precaution were a given, whereas in matters political those conditions might well be generated by our political decisions. When it comes to political matters, then, we do not just exercise caution: e.g. we don’t have planes landing on runways covered in ice. Nor merely take precautions as winter approaches: e.g. make sure airports have snowploughs, gritters and liberal quantities of de-icer standing by. We try, where we can, where it is policy outcomes that create the demand for caution and precaution, to exercise precaution in devising the policies that generate those outcomes.
What we mean is that, it might well be precautious to have snowploughs at airports, but if your airport’s exposure to the extreme weather is a result of your policy decision to locate it at altitude and in an exposed location, then you clearly should have thought precautiously earlier in the policy development process.
To take precautions is as much a part of being a practically rational actor, about being a person, as is being cautious. The argument in favour of making precaution explicit and central to policy formation is simply an argument against leaving it implicit and thereby likely overlooked.
And that is why a principled approach to precaution is wise. The Precautionary Principle takes seriously that the (relevant portion of the) past, our repository of evidence, is often pitifully small or unknown, or at best underknown compared to the totality of what could happen in the future. Or indeed, of what could have happened in the past, but (perhaps luckily) didn’t. Absence of evidence of harm in the historical record does not constitute evidence of absence of harm, nor evidence of no possibility of harm. This is especially so when the intervention in question has been tried only for a relatively short while). Or when the evidence concerning it is fragmentary.
A famous formulation of the Precautionary Principle is this, from the 1992 Rio Earth Summit: “Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation.” But the area of application of the Principle can be much wider than “the environment”: it applies, in principle, to any area whatsoever.
So, for example, in a case where we have no inductive evidence of harm and the complexity of the issue is such that deducing the precise threat of harm is difficult, we should invoke precaution.
2. EVIDENCE. The term “evidence-base” and derivatives have achieved the status of buzzwords. It is difficult to imagine a policy discussion or Powerpoint presentation related to matters of policy where there isn’t liberal mention of the term. A key reason for this is the rhetorical force this term now commands, owing to the success and influence of the Evidence-Based Medicine movement. This movement originates, in name at least, with the work of Gordon Guyatt and colleagues in Canada in the early-1990s, but has its methodological roots in the pioneering biostatistical work of the British physician and epidemiologist Archie Cochrane.
This success has led to the emergence of the Evidence-Based Practice movement, where the claim is that not only can all practices, in principle, be restructured to ensure they are pursued in a manner which is based firmly on good evidence, but that they must be so. The particular conception of evidence which is in play in the Evidence-Based Practice movement, of which evidence-based policy and politics are but instances, is often taken wholesale from the movement’s origins in medicine and specifically in the appraisal of drugs and other medical interventions in randomized-controlled trials (RCTs) and the meta-analysis of these.
The reason we mention this is to alert the reader to a potential problem when people invoke evidence-base in other contexts, such as policy, social work, clinical psychology, librarianship, and so on. The problem is that often the conception of evidence employed in the context of RCTs and their meta-analysis, statistical inductive evidence, is employed in domains in which it is less appropriate, or even unavailable in any meaningful sense.
Evidence is crucial, but what counts as evidence in different domains of inquiry is not a closed question. Very few historians or philosophers of science, much less practitioners in many of the “social sciences” would accept the term: the scientific method. Yet, the thought that there is the scientific method is precisely what it can seem like the Evidence-Based Practice movement is predicated on: that the same methods and rules for good evidence identification and gathering apply across a diverse array of domains of inquiry, and that these are at least akin to the methods honed in the domain of biomedicine, through the rise to prominence of biostatistical methods and RCTs.
In response to such methodological imperialism, it is often remarked that such methods are either difficult to obtain, or unavailable to domains of inquiry outside of the natural sciences, such as various areas of social inquiry. This remark often invokes notions of complexity: there are just so many variables. However, we would go further. It is not complexity that is the most telling problem when our attention is turned to human inquiry (humanities and social sciences), though to be sure there is a great deal of that. It is rather that “meaning” is central.
What someone is doing is a question about the meaning of their action: that meaning will perhaps draw upon that person’s narrative, and how they see that unfolding into the future, embedded in the practice they are currently pursuing, which, in turn, is embedded within a cultural milieu and specific socio-economic context, at a moment in the history of the society. RCT that!
Other methods are required. Ethnographic perhaps. Historico-comparative perhaps…
So, without becoming bogged-down in methodological discussion, we can make some general observations. In the context of political policy formation, to ask about evidence breaks down into asking the following two complementary questions:
a. Is there good evidence against the policy-intervention in question? Is there good evidence that it is damaging, risky, potentially-devastating, unpleasant, ugly?
b. Is there good evidence for the policy-intervention in question? Is there good evidence that it works? That its wanted effects are actual? That it is better than “placebo”?
Then, of course, what one needs to do is weigh up (a) against (b). Does one have evidence that likely/known “side-effects” outweigh known and desired, aimed-for effects? This assessment needs to be made judiciously, and without wishful thinking: mere hope that what one wants from the intervention will materialise is of course not evidence.
3. POLITICAL ECONOMY. We want to appraise the political grounds for giving the green light to a new policy. Let’s say genetically modified food crops. As those who read our piece on this topic (“What’s wrong with GM food?” TPM 65) will know, there are a number of ways one might pursue such an appraisal. In that piece we focused upon evidence, framing and precaution. But there is more – not just this parameter, political economy, but also the next one, asymmetry.
What is the “political economy” argument? Well, with regards to GM a good example might run as follows:
Genetic modifications can be, and are, patented. The seed patent then belongs to the company that developed the modification. Once in the environment the seeds spread, before long the wind has carried them far and wide and the crops of farmers who never signed-up for the programme contain the patented genetic modification. These farmers receive a visit from the patent holder’s lawyers, quickly followed by a large bill. Meet Corporation X.
Where these farmers’ agrarian craft once involved the preservation of seeds, year on year, preserving biodiversity and, through judicious selection passed down through generations, honing them to the environment, they now find themselves locked into the genetically modified seeds of Corporation X. In addition to having to pay royalties to Corporation X, these farmers must also purchase pesticide from it too, for that is the only pesticide that works with, i.e. doesn’t kill, these crops.
Is this outcome desirable? The political and economic control Corporation X now enjoys is extensive. Farmers who never chose to use Corporation X’s seed, even those who campaigned against its field trials and licensing, must pay for it, as migration and cross-fertilisation has led to them using it. The profits from these crops now become concentrated in Corporation X and farmers struggle to survive. Would we have made the decision to license these crops if we had worked out the likely political-economic consequences?
4. ASYMMETRY. Those who stand to benefit from or suffer from an intervention will generally, sooner or later, say or do something about it. Or at least, in theory, they have the opportunity to do so. But what about those who simply cannot do this? One should always think about the effect of an intervention on the very young, the severely disabled, the very old and infirm. And on non-human animals. And on untold unborn future generations.
Generally, the most significant of these groups are non-human animals and future generations, because these two groups are massively numerous – they number in the billions or perhaps trillions – and because they are profoundly neglected. Animals are neglected because of “speciesism”. Future people are neglected because we cannot see their faces, and because of the propensity, especially today, to a kind of chronic short-termism.
Let’s explore the case of future people a little more. One might, helpfully and following Edmund Burke, suggest that a people, seeking to determine what to do about some matter of policy, should be considered not as a present “time-slice” but as including past and future generations. Future generations matter more than past generations, because past generations are hard to harm. But it is only the present that gets to vote. We ought to seek to correct this systemic bias, this systemic exclusion from the people of those people who are most vulnerable: those who aren’t here yet.
One of us (Read) has suggested previously (“Guardians of the future”, TPM 57) that a systematic institutional solution needs to be found, to correct this bias. Following, but democratically updating Plato’s idea, Read proposes a “super-jury” of “guardians”, selected from the citizenry by lot, to have a role in making any decision that potentially impacts on the basic needs of future people. Similar institutions might be proposed to speak on behalf of the basic needs of non-human animals or the integrity of ecosystems.
Whenever there is a hidden, neglected or “silent” effect, either positive or negative, on a group that is voiceless, there is an “asymmetry”. i.e. those making the decision and those affected by the decision are not the same people. So that when there is something that would hurt the electorate but benefit the voiceless, it is (ceteris paribus) not advocated as strongly as it should be; and when there is something that would help the electorate but hurt the voiceless, it is (ceteris paribus) advocated more strongly than it should be.
This fourth parameter aims to correct this bias. We ought to foster ways of thinking that include the voiceless. It is the mark of a wise and caring polity that it takes this parameter seriously.
5. FRAMING. There has been an diagramme, a “meme”, doing the rounds on social media over the past few years; maybe you’ve seen it. We are unclear on its age and from whence it came. It shows a man operating an old-style large television camera, which is zoomed in on, and casting a shadow over, a part of a scene. The zooming, and therefore the framing of the scene by the camera, suggests to the viewer that what they are seeing, via the lens of the camera, is an image, cast in shadow, of an arm with knife in hand bearing down on another person, hand raised to protect themselves, mouth open in fear.
As one moves one’s gaze to the whole scene, beyond but including that part framed by the camera lens, one sees that what looked like an arm holding a knife was in fact the raised back leg of a running, fleeing, person, wearing, one has to assume, a somewhat pointy shoe. What appeared to be a face set in fear and an arm raised in defence, turns out to be a face bearing an expression of aggression, of someone in pursuit of the fleeing person.
The point the picture wants to make is that framing an image can change its meaning: in this case the aggressor becomes the victim, and vice versa. This is what framing can do, it can affect the meaning one sees in a picture. And while our discussion here has begun by reference to television pictures, or filmed footage, framing works linguistically too.
Linguistic framing works through the metaphors employed, and the emphasis one chooses to make, when representing an issue. Those who make a political decision to spend billions of tax dollars on explosives and employ those in an attack on another sovereign state and the people who live there, will frame this decision in a way that they hope will make it seem reasonable. They might employ their words with care, so as to frame military action in a certain way, designed to conjure up certain images.
Perhaps, therefore, they will focus talk on “precision guidance” and the “minimisation” of “collateral damage”, and not on the inherently indiscriminate nature of any explosive device and the inevitability of civilian deaths. Talk of precision serves to frame the attack differently to talk of large explosive blasts. Perhaps also they will talk of the “removal” of the regime or the political leader of the rogue state, rather than of an attack on a whole nation, on a sovereign state, designed to destroy its basic infrastructure and terrorise its civilian population.
In matters political we need to be cognisant, alert to, framing. Some frames, such as the rhetorical devices we touch on here by way of introduction, are quite crude, often the product of professional advisors on matters of political rhetoric (sometimes called spin doctors) and pervasive.
Other frames often reside in the subconscious and are operative through what we might have assumed were dead, comatose or cognitively-inert metaphors. Yet other frames are assumed, pervasive and seem to be bestowed upon us by the tradition or culture we inhabit.
Thinking of politics as primarily being about the freedom of individuals, the satisfaction of those individual’s wants, and the maximisation of want-satisfaction, might be considered a frame which is culturally-bestowed, having its origins and rational foundations in a particular socio-economic formation or moral tradition.
If one assumes what Charles Taylor referred to as “atomism” to be the frame through which one thinks about politics, then one will be constrained in certain ways, when thinking about what is politically desirable and possible.
Now, you might very well want to defend atomism, but that simply means that, along with the critics, you are willing to have the frame brought to the fore and made part of the picture, only unlike those critics you are not willing to move to alternative frames. Whether critic or defender, frames need to be made explicit.
So perhaps you’ve considered whether your proposed policy/intervention passes (1) the precautionary test, you’ve looked at (2) the evidence against and for it, at (3) how it stands to affect our political economy and our present society, and (4) at how it stands to affect those “outside” our present demos, the truly voiceless. But before you go ahead, you need to ask yourself one more thing: have you asked the right question? Were you addressing the right issue in the first place? Have you offered a solution to a non-problem? Is there a better way of looking at the whole thing? How should the whole thing be looked at in order for it to be comprehensible and motivating to a wider audience? These are the questions one asks when one asks about framing. Framing is crucial.
In conclusion, your thought might well be: How ought one to assess the five parameters? Are there rules for application? How might one tell whether what we’ve said so far in this article is right or not, whether it should be implemented?
We are not going to map out rules for this, in abstraction from case studies. This is only the beginning. We are, instead, going to produce some preliminary case studies. In the coming issues of TPM we are going take these five parameters as a guide to appraisal of actual, specific, policy issues.
For the next two years or so, you can look forward in each and every issue to our examination of one controversial policy or technology after another. We are going to look at cases including (but not limited to) human-triggered climate-change, nuclear power, nuclear weapons, antibiotics, antidepressants, expanding the franchise (e.g. lowering the age at which one can vote), and humanitarian military interventions (including the “Responsibility to Protect”).
We would welcome your suggestions for further cases for us to look at. We welcome strong disagreement with our arguments. We look forward to reader-produced Five Parameters Policy Case Studies, and we look forward to the next issue.