Dilemmas of Political Space

This section focusses on some of the issues of political space not necessarily related to the full formal model, but which have been raised in the spatial analyses of party politics. They are particularly relevant for the current work, which aims to show the benefits of thinking of political space not as a scientific but as a phenomenal space, something that should be studied and determined empirically and not assumed.

Inductive or a priori spaces

Although from the perspective of studying perceptual spaces, as outlined above, one could argue for a bottom-up inductive analysis of political space, it is still common to argue that this space is assumable and that there is merit in doing so. The distinction between an inductive and an a priori approach (Benoit and Laver 2006, pp. 50-52; Benoit and Laver 2012; De Vries and Marks 2012) to political space vaguely reflects the distinction between a phenomenal and a scientific space. In the latter case the political scientist assumes a certain number of dimensions with a certain content and then proceeds to locating parties on them and using these locations for conducting whatever analysis comes after. In the former case the political scientist does not assume anything about a particular

CEUeTDCollection

political space and strives to determine its characteristics from empirical data. If we would want to understand what is going on in the minds of people, to create a model which would capture or explain their behaviour, the preference should be for an inductive approach. The a priori approach has been argued for, however, for the reason that it saves us from potentially uncomfortable and problematic interpretations of spatial dimensions that can arise from inductive analyses (Benoit and Laver 2006; Benoit and Laver 2012).

Dimensions as ideology

If we talk about political space and its dimensions, we usually talk about ideology, something broad and general structuring the more specific elements of a political landscape, specific political issues.

This connection between issues and ideology is already present in the work of Downs (1957, pp. 97-98), who emphasises the distinction between policy and ideology. It is also present in later iterations of the general spatial theory, which assumes that ideological labels are used by people to guess the positions of parties or candidates on particular issues (Enelow and Hinich 1984). Ideologies are short-cuts that help voters so that they do not have to connect every single issue with their own world view (Downs 1957, p. 98), but they also must be related to specific policies if they are to reliably work as short-cuts (ibid., p. 102). An ideology should therefore be a generalisation of the political profile of a party, something that structures a party’s positions across many issues. Ideology or the dimensions of political space, although not directly seen, should in principle be observable if we look at parties’ positions on particular issues. Some issues appear together more and some less and this is how ideology should be empirically observable. However, if we look at the issue positions of parties (Albright 2010) in terms of how they covary, then it can be hard to see a simple or a general ideological structure.

Different spaces in different places

Political space can be different depending on where and how we look at it. It can be different for different actors, as well as in different places and different times. Not only is there ample evidence from empirical studies that the content and dimensionality of political space can change across countries and over time (e.g. Huber and Inglehart 1995; Benoit and Laver 2006), but it has also been suggested that the space, which exists in the minds of parties and the space that exists in the minds of the electorate are different.

Franzmann (2011, pp. 331-332) makes a distinction between programmatic heterogeneity and

CEUeTDCollection

ideological polarisation as two different kinds of divergence between parties that happen in different arenas. The first concept belongs to the supply side of politics, the parties, and refers to the total programmatic differences between parties. The second refers to how various issues relate to politically salient social cleavages and gives rise to ideological polarisation within the political system and the electorate more generally. It is thus more related to the demand side of politics. Such a distinction is also echoed by Stoll (2010b) and Stoll (2010a), who differentiates between a party and a voter defined space, which, depending on the level of aggregation of conflicts, either refer to issue spaces or ideological spaces.

Most empirical research does not effectively recognise this distinction, although it has been noted that data on party differences gathered from different kinds of sources correlate with each other to varying degrees. For example Dalton and McAllister (2014, p. 776) show how mass, expert and self-placement sources about party positions all correlate highly among each other, while data from party manifestos shows much lower correlations. This can indicate that there is a difference between how people think of parties and what is contained in party manifestos, although it has also been suspected to be due to measurement error (ibid.). The fact that different sources of data might be measuring different things has also been suggested by Meyer (2013, p. 31).

Ideological dimensions and the number of parties

Furthermore, empirically speaking, the nature of political space is related to the number of parties that inhabit that space, as well as the broader social context. If we try to measure party differences and how ideological dimensions manifest in the latter, we see, somewhat counter-intuitively, that the number of ideological dimensions that can structure empirical political spaces is dependent on the number of parties in the system. The latter determines how dimensions of ideology align if parties want to differentiate themselves from each other on them. There can be a range of possible positions on issues, but in the end for us to see a position, somebody, a party, has to take a position on it.

Taagepera and Grofman (1985) have argued that the number of parties in a system is related to the number of issue dimensions in that party system, an association echoed earlier and later in such works as Lijphart (1984) and Lijphart (2012). This ties the ideological dimensions in a society – issue cleavages – to the number of parties. In general, this is simply the acknowledgement that the structure of party systems reflects the issue structure – the conceptual space for political differences – in a society. The logic of this argument is that a certain issue structure in a society “generates”

a certain number of parties and Taagepera and Grofman (1985, p. 350) speculate that the choice

CEUeTDCollection

of an electoral system, that can accommodate more or less parties, might also be influenced by this environment. As they put it – parties minus issues equals one.

But there is another way that the number of parties and the number of dimensions is related, which is evident if we take a more data-centred perspective on the problem. If we talk strictly about how parties differentiate among themselves, then empirically speaking the first question should be, how can we see this difference? And here it should be kept in mind, that the dimensions on which parties are differentiated depends on the number of parties that we are looking at. Stoll (2010a) elaborates on this important point about the structure of political spaces being related to the number of parties that inhabit them. She emphasises that if we think from the perspective of party positions, then the dimensionality of political space is “the number of salient conflicts that are linearly independent given party positions on those conflicts, where two conflicts are independent or orthogonal if parties’

positions on one cannot be predicted using their positions on the other” (ibid., p. 409). She calls the latter the effective dimensionality.

This effective dimensionality is dependent on the number of parties that inhabit the space, as the differences between n objects can always be perfectly represented on at most n−1 dimensions (see section 2.1.2). Thus, a space that is needed to represent the differences between parties cannot have an effective dimensionality that is greater than the number of parties minus 1. This reflects the conclusion of Taagepera and Grofman (1985) noted above. In the case of two parties, the difference between them can be represented on a line – a single dimension.⁵ Whatever the number of issues that the parties differentiate themselves on, one side will always be associated with one party and the other side with the other party (or else there is no differentiation). All issues are aligned in the same direction and so there can be only one “visible” dimension of ideology where one party is on one side and the other party is on the other. In a similar way, all the possible differences between three parties can be perfectly represented on at most two dimensions and so forth.