Politics is sometimes described as ‘the art of the possible’[1] and so it would seem that distinguishing the possible from the impossible, the feasible from the infeasible, might be central to the study of politics. While there is now a lively debate around the role of feasibility in political theory,[2] much mainstream normative political theory and philosophy rather ignores this issue by focussing on ‘ideal theory’. An initial question then is, what is ‘ideal theory’ and how might it relate to real world politics? Our first difficulty in answering this question is that there is no agreement on any particular definition of the distinction between ‘ideal theory’ and ‘non-ideal theory’.[3] In a recent article[4] Zofia Stemplowska and I make a distinction between two forms of theorising that are sometimes conflated under the heading of ‘ideal theory’.  The first is essentially an exercise in conceptual clarification and analysis in attempting to set out a detailed account of a political ideal, such as ‘equality’ or ‘liberty’, we term this exercise ‘the theory of ideals’ (which might also include the study of the metrics by which we might measure deviations from an ideal and the study of how ideals might be combined to inform ‘all things considered’ evaluations). The second is then the discussion of the design of the institutions and practices that might realise a particular ideal in ideal circumstances – here  we  re-use the term ‘ideal theory’, which we place at an extreme of  a multi-dimensional continuum with non-ideal theory differing from ideal theory in that while ideal theory relates to institutional design under ideal circumstances, non-ideal theory relates to institutional design in non-ideal circumstances, where circumstances can depart from the ideal in many different ways.

Rather than focusing on the ideal/non-ideal distinction here, I propose to focus directly on the issue of feasibility – and draw out some of the issues relating to the role of feasibility considerations in political analysis.  Is a world free of nuclear weapons feasible? Is it feasible that the life expectancy at birth of all children be roughly equal, regardless of where in the world they are born? Is peace in the Middle East feasible? In asking these questions I do not challenge (or necessarily endorse) that each of these things may be desirable, nor do I intend to raise questions about the precise understanding of the examples cited (what exactly do I mean by ‘peace’, what exactly are the limits of the ‘Middle East); rather, I want to draw attention to potential limitations placed on political deliberation by considerations of feasibility, and by alternative senses of feasibility.

At first glance feasibility might seem a simple matter.  Either something is feasible or it is not, so that feasibility refers to a categorical distinction, and the boundary between feasibility and infeasibility is defined by a simple notion of possibility. If something is possible (no matter how low its probability) it is feasible. But first glances are rarely satisfactory, and one major part of the intention behind this note is to move towards a more complex, but more helpful approach to feasibility. In sketching a set of alternative approaches to feasibility we might start with the technical and work towards the political, with some detours along the way.

Technical Feasibility

A reasonably close approximation to our first glance understanding of feasibility might be gleaned from adopting a narrow ‘technical’ or ‘scientific’ focus: X is feasible if and only if it does not involve any breach of the true laws of science. Of course, this begs the question of whether science has ‘true laws’ but I will proceed without opening that debate.  On this account of feasibility the clear answer to all three of our opening questions is ‘yes’, but this is so only because the questions do not really make any demands on the laws of science, which suggests that the notion of technical feasibility is not going to be of much help in addressing most issues of interest in politics. But there are several further points to be made before we dismiss the idea of technical feasibility.

One point concerns the epistemic distinction between truth and belief, and how this bears on technical feasibility and possibility over time. Consider first the case of an earlier (or hypothetical) world in which everyone believes the world to be flat. In this case everyone would naturally believe sailing around the world to be infeasible, but they would be wrong in this belief. So, in cases of this sort, beliefs about feasibility (however widely shared) may come apart from actual feasibility, even where feasibility is defined in narrowly technical terms. Of course this also implies that beliefs about feasibility may change over time, as beliefs about science shift. So although technical feasibility is, by definition, a constant over time and place, even genuinely and widely held beliefs about technical feasibility may differ from actual feasibility and may change over time.

The second point is that the idea of technical feasibility may have more bite when we consider the joint feasibility of two or more X’s, rather than the simple feasibility of a single X. The point here is that technical feasibility may constrain us in slightly more subtle ways by ruling out certain combinations of events or actions, even if each action or event is individually feasible. In the simplest case even if it is clearly feasible from me to go to the cinema at 8pm today, and equally feasible for me to stay at home at this time, it is not technically feasible for me to be in two places at the same time – so that the combination of staying at home and going to the cinema is infeasible. It is often through this sort of combination effect that feasibility constraints bind – in effect forcing a choice between two or more alternatives each of which is feasible but which are jointly infeasible.

A special case of the issue of joint feasibility is the case of path dependence. In this case the time structure of events may influence what is feasible at some particular point. For example imagine that at some time T₀  both X₀ and Y₀ are individually feasible but jointly infeasible (so that they are strict alternatives), and further imagine that at some later time T₁,  X₁  is available if and only if X₀ occurred at T₀. Seen from time T₁ whether X₁ is feasible or not depends on which path was taken at time T₀. More generally, the set of feasible options at time T₁ will depend on the path of history up to T₁. Of course, from a timeless perspective, all coherent paths remain feasible and only incoherent paths are infeasible, but seen from any particular point in time, feasibility is constrained to a particular set of paths.

We will return to these points concerning the epistemic status of feasibility issues, the distinction between simple feasibility and joint feasibility, and path dependence below.

Human Feasibility

A second approach to feasibility would be to focus on what is feasible for us – as individual human beings. Of course, in a limiting case such a focus might lead us back to technical feasibility, if we assume that humans can do anything that is technically possible (where the relevant scientific limits include biological limits), but the basic idea here is to suggest that we might think that some things are rendered infeasible for us not on technical or biological grounds but because of other features of our status as humans.  This opens up the debate on human nature and, in particular the questions of the demands that we can place on humans. Is it feasible to expect individuals to treat all others as equals and so reject any special claims of kinship or friendship? Is it feasible to expect individuals to accept very significant personal sacrifice in the name of a political ideal?

To revert back to the discussion on ideal theory, one key element of many conceptions of ideal theory[5] is the idea of ‘full compliance’; which might be summarised as the assumption that (almost) all individuals are motivated to achieve the relevant political ideal. For example, is elaborating the political structure of a just society (under some particular specification of ‘justice’) an ideal theorist assumes that (almost) all individuals in the relevant society share a motivation to act justly and seek justice. By contrast, a non-ideal theory based on the same underlying theory of justice, might assume that human motivations are more varied and may include self-interest. Clearly, the task of designing political institutions that are maximally ‘just’ (in the agreed sense of justice) may be very different in the ideal world of full compliance and the non-ideal world of, at best, partial compliance.

So, an important basic question here is, does human motivation (or human nature more generally) form a feasibility constraint? Should we view humans as they are and accept that it is infeasible to change them, so that we should consider political and social institutions that work with the grain of human nature as it is, or should we view human nature itself as open to change via political means?

The form of this question almost certainly suggests a false dichotomy. The middle ground – in which we might distinguish between some aspects of ‘human nature’ that are basic and some that are more politically and socially constructed (and so variable) seems more attractive, but this only shifts the problem to that of identifying what  are the feasible and allowable means of changing human motivations or other aspects of human nature?

Political Feasibility

Moving from the individual to the collective, social or political is clearly a key step in any political analysis, and so the next issue is whether any further feasibility issues are raised by taking that step.

When we speak of political feasibility in our everyday political debates, we surely do not depend on the idea of technical feasibility, but rather we depend on an institutionally and historically rich account of feasibility that includes elements of human feasibility but also incorporates the ideas of joint feasibility and path dependence. A particular action or outcome may be politically infeasible just because the prevailing political institutions render that action or outcome infeasible – so that the infeasibility is not a property of the action or outcome itself, but rather of our current political system (including the role of individuals within that system). It might be the structure of the UN Security Council, or the nature of party competition in the UK, or the properties of a particular voting system, or any number of other aspects of our overall political system taken singly or in combination, that lies behind our diagnosis of political infeasibility.  And, of course, this implies that our judgements of political feasibility are likely to be contingent rather than absolute, in the sense that if the system changes, the pattern of feasibility and infeasibility changes.

In this sense then, each particular specification of a political system carries with it a pattern of feasible and infeasible actions, policies and outcomes. And it is here that issue of joint feasibility and path dependence may be particularly important. On the one hand, political systems and structures have to be seen as providing alternative packages of jointly feasible political outcomes so that an appropriate analysis would be to compare packages at the system level rather than to focus on the feasibility of specific policies or outcomes. On the other hand political systems and structures are themselves subject to complex path dependency, so that given a particular starting point at a particular point in time and with a particular set of political institutions in place, the feasibility of moving to another system, and so rendering a different package of policies and outcomes feasible, will itself be constrained.

For many issues – including the abolition of nuclear weapons, the equalisation of life expectancy at birth and peace in the Middle East – prevailing political institutions and systems may be seen as part of the problem rather than part of the solution in the sense that the historical and institutional setting in which these problems are located is such as to render their resolution politically infeasible. To the extent that prevailing political structures lock us into particular patterns of feasibility and infeasibility in the arena of specific policies and issues, the clear suggestion is that one has to see more major institutional and political reform as the key. But such a key carries dangers as well as opportunities. In shifting from one pattern of feasibility to another we may generally expect there to be losses as well as gains.  Nevertheless, the idea that we can think of political systems in terms of the patterns of feasible policies and outcomes that they support – rather than the particular policies and outcomes that may actually arise – provides a valuable additional perspective on the question of institutional design.

Conclusion Overview

The issue of political feasibility is a complex one and I have only scratched the surface in this brief note. The question of feasibility can be raised at a variety of levels and in a variety of ways, but I have suggested that initially separating out issues of technical feasibility, human feasibility and political feasibility can be useful, and that seeing political feasibility as a property of political systems and institutions with the emphasis on patterns of joint feasibility and path dependence can be helpful in providing a basis for discussions of political reform that put realistic questions of feasibility at the heart of the analysis. There is a sharp contrast here with the fundamental idea of ideal theory – which tends to sideline all issues of feasibility in order to focus on the question of desirability.

—

Alan Hamlin is Professor of Political Theory at the University of Manchester and was previously Professor of Economics and Dean of the Faculty of Law Arts and Social Sciences at the University of Southampton. He has held visiting positions at the Australian National University, at several US universities and at All Souls College, Oxford.  He is the Editor of the journal Constitutional Political Economy. He has recently published, Alan Hamlin and Colin Jennings, ‘Expressive Political Behaviour: Foundations, Scope and Implications’, British Journal of Political Science , Volume 41  Issue 03  July 2011 , pp. 645 – 670 http://dx.doi.org/10.1017/S0007123411000020

References

Brennan, G., and P. Pettit (2005), The Feasibility Issue, in Frank Jackson and Michael Smith (ed), The Oxford Handbook of Contemporary Philosophy, Oxford: Oxford University Press, pp 258-279.

Farrelly, C. (2007), Justice in Ideal Theory: A Refutation, Political Studies 55, 844-864.

Gilabert, P., and H. Lawford Smith (2012), Political Feasibility: A Conceptual Exploration, Political Studies.

Hamlin, Alan, and Zofia Stemplowska (2012), Theory, Ideal Theory and the Theory of Ideals, Political Studies Review 10, 48-62.

Miller, D. (2008), Political Philosophy for Earthlings, in D. Loeopold and M. Stears (ed), Political Theory: Methods and Approaches, Oxford: Oxford University Press, pp 29-48.

Mills, C. W. (2005), “Ideal Theory” as Ideology, Hypatia 20, 165-184.

Philp, M. (2010), What Is to Be Done? Political Theory and Political Realism, European Journal of Political Theory 9, 466-484.

Rawls, J (1971), A Theory of Justice, Oxford: Oxford University Press.

Robeyns, I. (2008), Ideal Theory in Theory and Practice, Social Theory and Practice 34.

Simmons, A. J. (2010), Ideal and Non-Ideal Theory, Philosophy & Public Affairs 38, 5-36.

Stemplowska, Z. (2008), What’s Ideal About Ideal Theory?, Social Theory and Practice 34, 319-340.

---

Further Reading on E-International Relations

• Constant Crisis and Political Byzantinism
• Beyond Rawls: The Principle of Transgenerational Equity
• Political Topology in the Anthropocene: Reconsidering Political Space in Light of Covid-19
• The ‘Erasure of Nationalism’ and International Relations
• Adorno on Late Modernity and Unfreedom: Reflections in a Global Era
• Re-Worlding China: Notorious Tianxia, Critical Relationality