1 Dimensions of political difference
Talking about the “positions” of political actors
Most people who talk about politics are likely to talk sooner or later about the “positions” of political actors. It is difficult if not impossible to have a serious discussion about the substance of real politics without referring to “where” key actors stand on substantive matters at issue.
The very notion of position implies distance. It is effectively impossible for any observer of real politics to describe the positions of two key actors without making at least an implicit statement that these positions are either “the same” or “different.” If they are different, it is difficult not to have some intuitive sense of whether they are somewhat different or very different. This intuition can become more systematic when describing the positions of three or more actors. Now, it is possible to make substantively meaningful statements such as “Churchill and Roosevelt are closer together on this matter than are Churchill and Stalin.” This is not a technical statement emerging from some formal model of politics. It is a statement easily understood by ordinary decent human beings who wouldn’t recognize a political scientist if they were mugged by one in broad daylight.
The very notion of distance implies movement. If my position differs from yours then it is conceivable these positions could move closer together: or further apart. It is also very common when talking about the ebb and flow of real politics to talk about people “changing” their positions on some important matter, with the result that they are now “closer to” or “farther away from” some other person than they were before. Once again, this is part of a common language people use when they talk about politics. Indeed, most political debate has to do with some people trying to change the positions of others on important matters at issue.
The very notion of movement implies direction. If my position moves closer to yours on some matter at issue, I have moved “towards” you on that matter.
All movement is relative. I can only observe and describe your movement relative to some benchmark. For example, it seems to be an uncontroversial part of the received wisdom of recent British politics that the Labour Party under the leadership of Tony Blair moved “towards the center,” and “away” from the more “left-wing” position it had occupied under the leadership of Michael Foot and Neil Kinnock. This statement is understood by everyone who is interested in British politics, political scientist or civilian, whether or not they actually agree with it. It is a statement that benchmarks changes in the positions of well-known British politicians against a commonly understood “left–right” scale, to which we shortly return, that has been found useful over the years for describing people’s positions on a range of important matters.
All of this goes to show that it is very common to think and to speak about politics in positional terms. Indeed it is very difficult to analyze real political debates without using positional language and reasoning. A skeptical reader should try doing this systematically for a sustained period of several months. It is not that we are making a helpful analogy between politics and the physical space within which we all live out our daily lives. It is much more than that. Most people – including those who are blissfully unaware of the mysteries of political science, as well as those who are utterly dismissive of them – find it difficult to talk about real politics in tooth and claw without using the notions of position, distance, and movement on the important matters at issue. These notions thus seem to have deep roots in the ways that people, from many different walks of life, think about and describe politics.
Positional political imagery has a long history, conventionally traced back at least to the Constituent Assembly that came into being after the French Revolution. The potential for chaos arising from the different “beliefs and wishes” of members of the Assembly in July 1789 is famously described by Thomas Carlyle (1871):
there are Twelve Hundred miscellaneous individuals; not a unit of whom but has his own thinking apparatus, his own speaking apparatus! In every unit of them there is some belief and wish, different for each . . . Twelve Hundred separate Forces, yoked miscellaneously to any object, miscellaneously to all sides of it; and bidden to pull for life!
(pp. 188–189)
But Carlyle goes on quickly (for him) to describe the emergence of order in the Assembly in explicitly spatial terms:
Nevertheless, as in all human Assemblages, like does begin to arrange itself to like; . . . There is a Right Side (Coté Droit), a Left Side (Coté Gauche); sitting on M. le President’s right hand, or on his left: the Coté Droit conservative; the Coté Gauche destructive. Intermediate is Anglo-maniac Constitutionalism, or Two-Chamber Royalism.
(p.192)
We return later in this book to the potential substantive meanings of a general “left–right” dimension, like the one used by Carlyle, as a benchmark for describing the positions of political actors. What is interesting in the present context is that similarities and differences in the positions of key actors are not only easily characterized in positional terms, but actually manifested themselves physically in the seating arrangements of the Assembly. Without going into the fine detail of “Anglo-maniac Constitutionalism,” furthermore, we find Carlyle defining an “intermediate” position between what he sees as two poles on the Coté Gauche and Coté Droit. Carlyle thus gives us a clear sense of a scale or dimension that runs from left to right. Certainly since 1789, the fundamentally positional notions of “left” and of “right,” with some underlying dimension defining intermediate positions, have been an important part of the common lexicon of politics. These notions have clearly been found useful as ways of conveying significant information about the motives and behavior of political actors.
A single “left–right” dimension, however, is not always enough to convey even the big picture. A clear example can be found when we set out to distinguish the positions of those who promote a “conservative” position on some matter from those who promote a “liberal” position – using liberal here in the “classical” sense associated with John Locke, Adam Smith, and, more recently, Friedrich von Hayek. If we are allowed only one descriptive dimension we can feel confident in saying that the conservative and classical liberal positions are both on the right; but we also are acutely aware that they are distinctly different; and we are not comfortable with describing the conservative position as being clearly to the left of classical liberalism, or vice versa. These two positions differ on some other dimension. At one end of this other dimension we find a quintessentially conservative belief in the value of loyalty to the nation and/or state, an organic view of the interrelationships between citizens and society, and possibly also a belief in the central role played by the church in binding citizens and society together.1 At the other end of this dimension we find a quintessentially individualistic liberal belief in the primacy of the citizen vis-à-vis the nation and/or state, a consequent suspicion of state intervention in the lives of individuals, and often also a very firm belief in the need to separate the roles of church and state.2 The resulting “liberal–conservative” dimension is another indispensable tool for describing the positions of different political actors. Without it, classical liberals and conservatives, who self-evidently promote different ideas, cannot be distinguished.
The richer the description of politics we seek, the more dimensions we need to describe the positions of political actors. The more dimensions we use, the more fine-grained our descriptions of politics can be. More dimensions are not always better however, since adding ever finer-grained detail does necessarily not make for ever-more useful descriptions of the world. When we set sail across the Atlantic, for example, we would get lost if our only charts were so detailed that they show the position of every single grain of sand on every single beach we might pass. We need a description of the (political) world rich enough for the purpose at hand, but not so rich we cannot see the beach for the grains of sand.
The bottom line in all of this is that most people, whether they realize it or not:
- almost certainly think about politics in positional terms;
- almost certainly need more than one dimension to describe important political interactions;
- typically use no more dimensions than they really need to describe the interactions in which they are interested;
- feel they share with others a common understanding of the meaning of these dimensions and associated terms such as left, right, liberal, conservative, and so on – so that conversations with others using these terms are meaningful.
All of this motivates a systematic attempt to describe politics in positional terms using a limited number of underlying dimensions. This has been reflected in the steady growth of interest by political scientists in “spatial” model of politics. Indeed, one of the profession’s most distinguished practitioners, Gary Cox (2001), recently described the spatial model as “the workhorse theory of modern legislative studies.”
If these spatial models are to be expressed in terms of real-world politics, rather than remaining as mathematical abstractions, it is necessary to measure the positions of key political actors, which is the core purpose of this book. Before rolling up our sleeves and getting down to the business of measuring policy positions, however, we must first consider some matters we have so far avoided. These have to do with what, more precisely, we mean by the “positions” of particular political actors if we abandon our current implicit perspective as God-like observers looking down on the political world from on high and get down to street level, looking at the world through the eyes of individual citizens.
Ideal points, policy “positions,” and policy “distances”
Things do look different at street level. We can be reasonably confident that each individual knows what he or she wants, more or less. (We won’t get into what the world would look like if nobody had the slightest clue about his or her own personal tastes.) But we also can be confident nobody knows for sure what anyone else wants. All any individual can do is draw inferences about the tastes of others from the systematic observation and analysis of their behavior. In the present context, we are interested in two particular aspects of these inferences. The first is that individuals are aware that the behavior of others is potentially “strategic.” Thus when other people tell you things, you must always consider whether or not they are telling the truth since you know they often have strong incentives to lie. The second problem is that the meaning of the dimensions used by a particular individual to describe the positions of others is subjective to that individual. This forces us to confront the possibility that the other people with whom we interact are navigating the political world using a map of this that differs radically, and in ways incomprehensible, from our own.
We could take a puritan line on this and conclude that the endless possibilities for strategic dissimulation, and the subjective nature of any individual’s map of the political world, mean that there is no possibility of finding a commonly understood set of dimensions we can use to describe and analyze real politics. This would imply that the use of positional language to describe politics might be very pervasive, but is ultimately misleading.
We are not puritans. We feel that very many, very enlightening, descriptions and analyses of politics have, over the generations, used spatial language and reasoning. We find it particularly compelling that intelligent and well-informed commentators – most of whom know nothing whatsoever, and care less, about political science and its “spatial models” – have coordinated over many years on positional language and reasoning as an effective way to communicate with each other. Thus the question for us becomes one of how positional language and reasoning have become so useful and pervasive, in the face of the serious epistemological and methodological problems that appear to confront us when we try to view the political world from street level.
“Sincere” or “strategic” positions?
If we can assume anything at all when we talk about politics, we can assume each individual to be motivated by a set of beliefs, needs, and desires that condition how s/he behaves in a given situation. Describe as “preferences” the set of desires that motivate a given individual in a particular context. Such preferences may be highly contingent on context and/or highly conditioned by the political process under observation. But, for a given point in space and time, there is a particular set of preferences that motivate each individual. We have to start somewhere and we start with this.
We also know these preferences are intensely private to the individual3 and effectively impossible even for fully socialized and articulate individuals to communicate perfectly to each other. There is always a potential for misunderstanding. Furthermore we know that, in most societies, part of being a fully socialized individual is not saying exactly what you think in a given situation. Indeed saying exactly what you think in every situation you find yourself in is considered a form of “disinhibition” and seen as a mild social and/or mental disorder. Add to this the widely recognized and pervasive incentives facing every individual to dissimulate for all sorts of reasons, which means that we can never be certain that any individual claiming to be sincerely communicating his/her preferences really is being sincere. We are left, whether looking at any other individual from a position as fellow players in the game or as disinterested external observers, with at least two different “positions” for that individual in relation to any matter at issue. There is a sincere or “ideal” position, which reflects the individual’s own beliefs, needs and desires. This motivates his or her actions but is fundamentally unobservable by others. And there is a “public” position, which can be inferred from a person’s words and deeds.
Our social and political interactions with others involve, among other things, anticipating their behavior in different circumstances on the basis of inferences we draw about their ideal positions from their public positions, viewed in strategic context. A person’s public position is all we can observe; everything else is a logical inference from our observations of others, using some explicit or implicit model of the situation under observation. Ordinary decent humans draw these inferences intuitively and informally. Experimental social scientists draw them on the basis of carefully controlled experiments. But what everyone is doing is using some particular model to draw inferences, from the observed actions and statements of other people, about how their fundamentally unobservable private preferences are likely to motivate their future behavior.
These inferences are not easy to draw in a systematic and rigorous way but they are thankfully not our concern in this book. Here, we are very explicitly dealing with the observed positions of important political actors, as these appear to groups of people who are skilled at observing these things, and who in effect we treat as God-like external observers looking at the political world from on high. It may well be, for example, that we feel strongly some political party is “really” more extreme than its stated position on some policy dimension under investigation – that its stated position does not reflect its true position. But that is a case to be made and investigated on the basis of some precise model of party competition, for which the position actually observed will be an important empirical referent. The estimated party positions we present later in this book are thus publicly observed positions – we make no claim whatsoever about what is “really” going on inside the heads of particular party politicians.
The subjectivity of spatial maps
There is a ring of solipsism to the superficially attractive argument that every one of us views the political world from his or her own unique perspective, so that no common view of the political world can be shared by a group of people who interact with each other. It is thus useful to consider whether we can extend to our present concerns the robust argument, following Wittgenstein, that solipsism is incoherent because constructing the solipsist argument requires using a language that intrinsically implies some shared understanding of certain words, thereby admitting the very premise the solipsist argument denies. In the present context this might take the form of arguing that, by accepting that the study of political interactions between two or more human beings is both possible and potentially fruitful, we accept the notion of interaction. This intrinsically involves political actors being able to anticipate the actions of others to some degree; this in turn implies they can see the world to some degree through the eyes of those others. Indeed humans who are deemed quite unable to see the world though the eyes of others are typically considered to be in some sense mentally disturbed and/or sociopathic. Poli...
