The Normative Value of Partisanship: When and Why Partisanship Matters

The Value of Partisanship

The study of partisanship has traditionally been at the core of political science, and recently, it has enjoyed a resurgence of interest among political theorists (Muirhead, 2006, 2011; Muirhead and Rosenblum, 2006; Rosenblum, 2008; White and Ypi, 2010, 2011, 2016). The driver for this resurgence is the view that citizens’ disaffection with political parties weakens the ability of citizens to influence policymaking, thus depriving institutions of the civic engagement necessary for democratic legitimacy (Offe, 2006; Schmitter, 1995; White and Ypi, 2010: 809). This article considers the normative role that partisanship might properly play in democratic politics. ¹ I will defend three conclusions about the value of that role. First, partisanship is not on a par with prejudice or partiality. I endorse the view that partisanship can have genuine moral content, but maintain that partisanship does not necessarily express a democratic virtue or any other particular normative principle. ² Second, the conception of partisanship which interests political theorists should emphasise the normative dimensions of party commitment. The aim here is to present a normative defence of partisanship that accounts for both its intrinsic and its instrumental importance for contemporary multi-party democracies. Third, a normative account of partisanship may not be able to do all the work that some political theorists want to assign to it. Partisanship is but one aspect of party politics and political commitment – and only under certain conditions is it ‘a seedbed of civic virtue’ (Putnam, 2000: 19, 152–154).

The structure of the argument proceeds as follows. In the next section, ‘Defining Partisanship’, I examine the concept of partisanship as used by political scientists and discuss why a normative account of partisanship, as a set of reasons for partisanship, needs to be congenial with the commitment citizens show to political parties. Then, in the section ‘Partisanship Revived’, I distinguish between two possible defences of partisanship: an intrinsic defence and an instrumental defence, and discuss recent attempts to support these positions. In the section ‘Partisanship as Commitment’, I outline the difficulties associated with an intrinsic defence of partisanship and support the claim that such a defence must be conditional on whether or not it expresses a particular motivational structure – one that is based on civic commitment and not just on party fidelity. The reason we consider partisanship normatively significant, I argue, is because we consider its actions to more successfully express, and nourish, a particular motivational structure than other actions. In the section ‘Partisanship as Contestation’, I sketch an instrumental defence of partisanship. I anchor this defence to the political contestation that commonly characterises partisanship and offer an account of the value of having such a conception available in democratic politics. I argue that commitment leads to contestation among both partisans and non-partisans, and that a world containing no contestation of political commitments would be worse than one in which there is such contestation. Finally, and to address a number of objections, I criticise arguments that have been offered in support of a more narrow conception of normative partisanship than the one I defend here.

Defining Partisanship

Partisanship is a highly operationalised concept in the study of political behaviour. Before I evaluate its normative significance, I will briefly examine how political scientists have understood the concept and how this understanding frames the adoption of the concept by political theorists. I take political scientists and political theorists to subscribe to the same definition of partisanship, ‘the political orientation of citizens who stand with a party. This orientation is both psychological in that citizens identify with a party and behavioural in that they usually vote for it, and possibly give it time and money’ (Muirhead, 2006: 714). In this article, I adopt a definition of partisanship which comes close to the contemporary understanding of the concept by placing commitment to a political party at its core. This conception of partisanship prioritises party loyalty (Mair, 2013: 35–37). This is important because, in the absence of a nuanced understanding of partisanship, the term either gets too inclusive or becomes too difficult to distinguish from the much broader category of party politics.

In providing an account of possible defences of partisanship, I think it is useful to look at how empirical political science understands the term and its role in democratic politics. The locus classicus for the study of partisanship by political scientists is the work of the so-called big four of the Michigan School (Campbell et al., 1980). These four ‘founding fathers’ considered partisanship to be an example of social identification. According to their view, the individual’s self-concept is derived from her knowledge of her membership of a group, like a political party, together with the value and emotional significance attached to that membership (Holmberg, 2007: 559). In these accounts, affect and group belonging is emphasised over cognitive factors. Party identification is seen as being formed mainly by socialisation in the first third of one’s life (Green et al., 2002). Hence, party identification is conceived as an exogenous variable, ‘affecting politics while not being affected by it’ (Holmberg, 2007: 564).

More recent work builds on the work of the Michigan School and its emphasis on party identification. But whereas the earlier work of the Michigan School emphasised that partisans deflect information which is inconsistent with their party attachments, more recent work concedes that partisans can change their perceptions of the parties and even vote for other parties, but, crucially, without changing ‘the team for which they cheer’ (Green et al., 2002: 8). Furthermore, recent research on partisan sorting has shown that voters’ partisan and ideological commitments are closely linked. This is due to party elites becoming more ideologically polarised. This elite-level phenomenon, it is argued, provides voters with information to understand the policy positions adopted by the two parties. Voters use these cues to align their partisan and ideological beliefs. As a result, partisans become more conservative or more liberal by adopting the ideological outlook of right or left party elites, respectively (Levendusky, 2009).

The deep-rooted nature of partisanship is confirmed by studies showing that political competition dramatically shapes the salience of partisanship in interpersonal trust (Carlin and Love, 2016). Evidence suggests that trust among co-partisans is high, whereas it is low among rival partisans. These inter-group trust gaps are greater than those based on social identities such as socioeconomic status. This makes it easier for voters to select parties which represent their interests, even though it undermines the ability of supporters of different parties to cooperate across party lines (Carlin and Love, 2016). Some studies even find that genetic factors strongly influence the decision to identify with any political party, and that certain personality traits influence the intensity of party identification (Gerber et al., 2012; Settle et al., 2009). Experimental studies on the influence of altruistic concerns on voting behaviour also suggest that partisan altruists are much more likely to vote than their non-partisan or egoist peers (Fowler, 2006).

This account of partisanship has been challenged by the so-called revisionists and in Fiorina’s work on the importance of cognitive factors and retrospective political evaluations as formative factors for party identification (Holmberg, 2007). This approach draws heavily from Rational Choice Theory (RCT hereafter) and economic theories of voting (Downs, 1957; Fiorina, 1981). Party identification, for this group of scholars, is a cumulative tally of retrospective and cognitive evaluations of the performance of a party in delivering what the voters want. It is seen, therefore, as an endogenous variable which changes (or remains constant) as a result of the retrospective evaluation of political parties’ performances (Clarke et al., 2009; Holmberg, 2007). Data analysis from a number of societies, including countries as different as Britain, Germany and the United States, suggests that partisanship has dynamic qualities that are sizeable and on-going. In these societies, researchers find not just large numbers of stayers who remain committed to a particular party over a long period of time but also large numbers of movers, which they argue is difficult to square with the Michigan School (Clarke et al., 2009). Hence, the main dispute between the two approaches is the extent to which partisanship must be treated as an exogenous or an endogenous variable. But what exactly are the implications of this research for the normative study of partisanship?

The two conflicting models discussed above allow for two different accounts of partisanship: a non-cognitive and a cognitive one. For the Michigan School conception of partisanship, the actor does not reflect on the motives but simply unconsciously follows his or her ingrained beliefs. In contrast, for rational choice conceptions, actors think about the private returns from supporting a party and are seen as unconcerned about public benefits. The potential problem here for normative political theory is that neither conception fits into a clearly defined moral framework – if by that one means individuals sacrificing their interests for the greater good – as neither model requires that the individual should pursue collective welfare in order to have a partisan attachment. I think this is an accurate description of the two models, but it misconstrues the relevant desiderata that a normative defence of partisanship must meet. To support the claim that Michigan and RCT are compatible with a moral defence of partisanship, one does not need to claim that both models require that an individual should pursue collective welfare. One only needs to show that at least one model (if not both) can explain, or account for, why an individual may want to pursue collective welfare.

Here is why this is the case: one’s commitment to a particular political party need not be the product of a deliberate choice in order to be a motivating reason and to have moral value. For example, if one is educated to give priority to one’s civic duties over brute self-interest, then this has moral value if it brings about better government, even if such behaviour is purely, or mainly, the result of socialisation. A moral ideal does not necessarily lose its moral worth by being socially conditioned. Therefore, the Michigan School, with its emphasis on socialisation, fits within the moral framework adopted in the article. RCT can be formulated in a broader way if one defines rationality as maximising one’s preference order. If one adopts this more inclusive definition of RCT, then one need not continue to associate rationality so closely with maximising one’s self-interest. A partisan’s most highly ranked preference could be a concern for the public good (as her party defines it) and therefore be moral in character. Since there is nothing irrational in wanting to maximise one’s preference order (regardless of its content), it follows that partisans who are motivated by a concern for the public good are not irrational. ³ Therefore, both the Michigan School and RCT fit in a moral framework.

Another potentially troubling aspect of partisanship, especially in terms of its normative defence, is that it is not statistically correlated with either conservative or progressive electoral results (Holmberg, 1994). Partisanship is un-partisan in its substance. Hence, defending partisanship, at least in the world as we know it, does not appear to be tantamount to defending conservative or progressive values. Therefore, if we are to defend partisanship, we need to defend it for what it contributes to civic debate and government and not on the basis of the substantive political results it delivers.

It is also important at this point to start distinguishing between the normative and empirical dimensions of an account of partisanship. Normative political theorists are interested in whether partisanship exhibits normatively valuable qualities, whereas empirical approaches are primarily concerned with examining the causal determinants of partisan behaviour.

There is also a difference between a normative assessment and a normative defence of partisanship. A normative assessment of partisanship need not be empirical in the first instance. It suffices to list normatively valuable principles or ideals and then to argue that partisanship needs to have qualities that either express or serve those principles and ideals in order to have normative value. A normative defence of partisanship could then follow in the form of a positive normative assessment of partisanship which aims to show that partisanship has, or could have, such qualities.

There is then a significant degree of overlap between empirical and normative approaches. If partisanship were to lack the capacity to exhibit the normative qualities political theorists deem valuable, both under current and favourable circumstances, then partisanship would simply lack the normative qualities which will render it interesting for political theorists. The fact that partisanship does not always empirically exhibit those normative qualities, or that it does so only to a certain extent in our current predicament, only sets feasibility constraints to normative expectations from partisanship. It does not reduce the normative value of the qualities that partisanship needs to have, or serve, to have normative value. Empirical work can play a crucial role here, nonetheless, by adopting a forward-looking approach that aims to identify the relevant feasibility sets in a variety of different conditions and, hence, the degree to which partisanship has (or could have) normative value in those conditions, and also by acting as a useful precaution against idealising generalisations and unrealistic expectations. Without a normative compass, however, empirical research runs its own risks: it lacks direction and is in greater danger of succumbing to status quo bias. ⁴

These are the wider methodological parameters within which I develop an intrinsic and instrumental normative defence of partisanship in the sections ‘Partisanship as Commitment’ and ‘Partisanship as Contestation’, respectively. The next section provides a summary of normative approaches to partisanship, which together with this section prepares for the analysis that follows.

Partisanship Revived

Normative defences of partisanship can be categorised into two broad types. First, there is a category of arguments which emphasise the ‘moral distinctiveness’ of partisanship. The moral qualities that are emphasised here vary to some extent but they are clustered around certain central themes that are seen as typical of partisan activity: commitment and loyalty to a political project or to democratic legitimacy more generally, a sense of community and friendship fostered by acting together and identifying strongly with those who share similar conceptions of the common good, and finally an ability to advance one’s cause with civility as well as the willingness to compromise with one’s rivals or opponents (Muirhead, 2014: Chapters 1–3, 5; Rosenblum, 2008: Part III, esp. 341–366; White and Ypi, 2016: Chapters 4 and 5). We may call these arguments intrinsic defences of partisanship, as in these accounts the normative value of partisanship is treated as being inherent to the practice. ⁵ Second, there is another category of arguments which underline the important and creative role that partisanship has in serving and improving civic debate, or even justice. ⁶ The emphasis here is on the role that partisan contestation plays in political argument, as well as on the role of partisans in supplying substantive reasons, and in that sense valuable information, to the citizenry as a whole. ⁷ We may label this second group of arguments instrumental as they aim to highlight the constitutive role that partisanship has in improving democratic and just government (Muirhead, 2014: Chapter 4; Rosenblum, 2008: 139–148; White and Ypi, 2016: Chapters 1 and 3).

These two categories of arguments are vulnerable to three objections. The first concerns the normative foundations of partisanship. It is far from clear how giving normative value to partisanship can be easily reconciled with approaches in political theory which emphasise the importance of ideals such as reflexivity (Guttmann and Thompson, 1996) and impartiality (Rawls, 1999) in providing the relevant justifications (Bonotti, 2014). Defenders of partisanship appear divided on the extent to which partisanship is consistent with prevailing theories of democracy and justice. This is particularly relevant to instrumental defences of partisanship because if partisanship is to have normative value, it needs to be demonstrably able to serve these normative ideals. In the penultimate section, I examine how partisanship can enhance these ideals.

A second set of problems with normative appraisals of partisanship concerns normative content. The relation between partisanship and a sense of commitment to a community with a particular political outlook is often emphasised here, along with a sense of civility and compromise. But the moral content of partisanship in these accounts is often both too wide and too narrow. It is wide because it gives the impression that the value of partisanship depends on an often overtly inflated duty of civility rather than a prudent appetite for compromise (see Muirhead, 2006, 2014). It is narrow because a sense of commitment to a community with a particular political outlook is deemed to be good, even when it is not motivated by a civic ethos centred on a concern for the public good (see Rosenblum, 2008). In the section ‘Partisanship as Commitment’ below, I sketch the content and direction of such a civic ethos and its role in an intrinsic defence of partisanship.

Further problems concern the normative scope of partisanship. Political parties, despite their significant role in maintaining and cultivating a civic ethos, are neither capable of nor appropriate for carrying the full burden of such a heavy theoretical bandwagon. Recognising their role in promoting a civic ethos that reinvigorates democratic politics must not lead to their idealisation and the colonisation of civic society by party structures (Habermas, 1998: 476). This is because parties serve not only as servants of civic virtue but also as institutional incubators of political vice (Mair, 2013: 20). Hence, we need to guard against the tendency to inflate the concept of partisanship in order to serve purposes that can be only tentatively, contingently and indirectly served by it. The distinction between classic and normative, as opposed to modern and empirical, accounts of partisanship is therefore unnecessarily divisive (White and Ypi, 2010, 2016). The challenge for political theorists is to provide a coherent and clear set of criteria for its normative appraisal that combines the insights of both approaches.

Partisanship as Commitment

Before providing an instrumental defence of partisanship and an account of how the contestation it fuels serves the public good, it is important to examine whether one can construe an intrinsic defence of partisanship. Supporters of partisanship who proceed in this way need to show that partisanship has properties that are inherently normatively valuable. These properties of partisanship obtain their value either as an instantiation of ideals such as democracy and justice or as an expression of competing but important values that the ideals of democracy of justice need to accommodate.

Let us start by examining some of these arguments (Muirhead, 2006; Rosenblum, 2008). The first step in providing an argument as to the intrinsic value of partisanship is to provide a definition of the relevant ‘positive effects’, or value, to which partisanship is directly anchored. Since I have defined partisanship as strong identification with a political party, any positive effects need to be linked in some way to the constitutive elements of that definition. Take the case of the democratic ideal. For the argument to work, the central role of political parties in democratic politics needs to be shown to express democratic virtues such as democratic legitimacy. For instance, a statement in support of such an argument would be that ‘the fundamental thing that parties do is to manufacture majorities that constitute the collective binding rule’ (Muirhead, 2006: 718). The argument here in support of the legitimacy of the majoritarian constitution of collective binding rules to which parties are intrinsically linked is that ‘the majority have a kind of moral claim to legitimacy simply by their numbers. It should be that way because aggregation is a necessary way of recognising the equality of citizens who under conditions of freedom will inevitably disagree’ (Muirhead, 2006: 718).

I accept these points. Nonetheless, they are all points about the moral legitimacy of majority rule and the instrumental role parties play in the function of majority rule in representative democracies. They are not arguments about the moral significance of partisanship as such. To achieve the goal of an intrinsic defence of partisanship, we need to go a step further in linking the function of parties in majority rule with partisanship. Russell Muirhead attempts to do this when he claims that partisanship ‘expresses a willingness to make a good faith effort to stand with a group striving for democratic legitimacy’ and with a group that aspires to be ‘ideally the largest of groups’ (Muirhead, 2006: 719, 2014: 89). I take the phrase ‘good faith effort’ to mean that what is distinctive about partisans versus others (i.e. voters or persons of conscience) is this moral willingness. Partisans identify strongly with a party because they want to strive with others for that party’s democratic legitimacy. Thus, Muirhead’s argument can be broken down into the following propositions: (1) Majority rule bestows legitimacy on collective binding rules. ⁸ (2) Parties are necessary for the function of majority rule. (3) Partisans identify with parties’ aims to serve that function by aspiring to become majorities; they want their party to become the majority. (4) Therefore, partisans are doing something good for democratic legitimacy.

To present this argument in the context of the discussion in this article, let me give an example that, by challenging the above conclusion, will I hope clarify what is at stake. I start from the assumption that citizens have a duty to support democratic virtues, and that partisanship is a necessary means to fulfil this duty in a representative democracy. I then describe certain hypothetical cases in order to examine the reasons which motivate citizens to become partisans, and to thereby examine the validity of the above argument in support of the intrinsic defence.

Take the case of Citizen K., a media magnate: his partisanship seems to be motivated not by his moral duty to support democratic virtues but by his belief that his preferred party will pursue policies which will favour his media enterprises. Citizen K. would in fact love to see his party become a majority. Regardless of whether his belief is true or false, my view is that Citizen K. is partisan for the wrong reason; he is clearly not motivated by a moral appreciation of the ideal of democratic legitimacy. It seems to me therefore that a normative defence of partisanship should not be intrinsically reliant on conforming to democratic legitimacy, if only because of the plurality of reasons underlying partisan behaviours. Even if parties are necessary for the function of majority rule, it does not follow that the motivation to support those parties is the result of identification with the ideal of democratic legitimacy.

It follows from the case of Citizen K. that those who identify strongly with a party and its hope for a majority are not necessarily driven by a ‘good faith effort’. But one could object here that Citizen K. is not a true partisan. Citizen K. is driven by his self-interest and not a moral striving for democratic legitimacy, understood as legitimacy by large numbers. But if Citizen K. is not a partisan, then strong identification with a party is not tantamount to partisanship; it may indicate a striving for democratic legitimacy, but one which does not involve a ‘good faith effort’. Hence, legitimacy by large numbers does not require the good intentions of partisans.

The broader conclusion that one can draw from the illustration above is that not all partisans identify with democratic legitimacy. Therefore, partisanship does not necessarily express a good faith effort to bring about good effects such as democratic legitimacy. If strong but self-serving identification with a party is not a case of partisanship, because it is not intended on the basis of democratic legitimacy, the definition of partisanship starts to differ significantly from that espoused by many political scientists. ‘Partisan’ identification with a party which aims at a majority becomes illegitimate in the moralised version of the definition because it is based on an individual’s non-beneficent reasons.

Avoiding this problem of shifting definitions comes at the price of redefining partisanship in a way that is still congenial to democratic legitimacy but highly idiosyncratic. It entails denying that one can strongly identify with democratic legitimacy (for instance, with the principle of majority rule) without identifying strongly with a specific party or even with a party system. It also entails denying that majority rule confers legitimacy on a political decision in the absence of a system of political parties and therefore in the absence of partisanship. ⁹ Partisanship is distinct from party preference, support for a party system, or democracy in general. Therefore, those who confuse strong commitment to a political party with strong support for a party system or democracy are committing a category mistake. ¹⁰ Drawing an intrinsic link between partisanship and substantive values like democratic legitimacy does not directly follow from what we typically understand as partisanship. ¹¹

It seems, therefore, more appropriate to adopt a more qualified defence of partisanship. This can be mounted by adopting a civic conception of democratic legitimacy, that is, the status enjoyed by political institutions to the extent that they deliver good government by servicing the public good rather than factional interests. ¹² If partisanship is defined as the commitment to a particular political institution, then to have normative value the political party must intrinsically express a concern for the public good.

The commitment characteristic of partisanship is key to such an intrinsic defence of partisanship. It is also at the heart of many of its defences, which are examined in this section. This is, remember, the view that partisanship is an expression of something which is inherently normatively valuable. In what follows I discuss what I consider to be the most promising version of this kind of defence and explain why it matters for a normative defence of partisanship, as well as why it is insufficient for an intrinsic defence of partisanship. More specifically, supporters of this view argue that a society without political commitment is one that is lacking an important property that is normatively valuable. It is a society of disengaged, excluded and disaffected citizens who are subjected to rules without contributing to their making (White and Ypi, 2016: 77). The argument is that partisanship consolidates and promotes political commitment. It provides not just organisational structures which nourish and support political commitment but also motivational and epistemic resources. ¹³ In that sense partisanship expresses an ideal of political community to which people are strongly dedicated, by making widely available a set of justifications and policies which support political commitment.

This ‘thick’ depiction of partisanship helps to explain why citizens are willing to engage in partisan activities even when those activities entail significant participation costs. As individuals are willing to endure costs due to their commitment to a friendship or their teammates, so too for parties to which they are committed (White and Ypi, 2016). I agree with these points. There is nothing particularly mysterious or irrational in paying a price for sustaining and promoting a collective project one considers worthy. ¹⁴ But the emphasis on commitment alone is insufficient for a robust normative defence of partisanship. Partisan commitment could be directed not just to identify and challenge imbalances of power and factional interests but also to disguise and support them. It can be rallied to support not just the collective self-rule of the many but also collective subordination to the few. Therefore, partisan commitment needs to be coupled with a particular sense of moral direction to express a concern for the public good.

Here, I attempt to construe an intrinsic defence of partisanship. In doing so I claim that partisanship must reflect a moral motivation in order to have moral value. Acting from duty is acting from the belief that one’s act is morally required. It is to ϕ because (one believes) ϕ-ing is right. This conception of the moral worth of an action is congenial to a possible intrinsic defence of partisanship. ¹⁵ I believe that the same distinction applies to partisanship. The reason why we, in some instances, consider partisanship intrinsically moral is because we consider partisanship to be an action which can express a motivational structure of ‘rightness’. But if we treat all partisans as acting out of right reasoning, a goodwill which gives priority consideration to civic duty over all other interests, we ignore the fact that not all acts of partisanship necessarily express that motivational structure. We should, therefore, consider only some cases of partisanship as having moral worth. It is partisanship’s motivational structure, and not its intrinsic meaning, that renders some partisanship intrinsically morally worthy.

Following this approach, partisanship can be viewed as an expression of civic virtue that cultivates the kind of commitment required for civic engagement. We can treat certain instances of partisanship as cases of acting according to our capacity to be moral persons (Rawls, 1999: 442) and, hence, as examples of a desire to act according to political norms which serve the common good, as opposed to merely our private ends. Drawing a distinction between partisan commitment and fidelity helps us identify those cases of partisanship which exhibit a truly civic motivational structure. Fidelity to a political party, unlike commitment, need not express this commitment to the public good. In this regard, ‘partisanship’ has intrinsic moral value when it expresses a commitment to acting out of a concern for the public good, whether or not one’s own welfare is affected by it, whereas ‘fidelity’ is concerned with undertaking the very same actions but only to the extent to which one’s own welfare is affected (Nagel, 1970; Sen, 1977). ¹⁶ The basic tenet of this conception of partisanship is that a democratic and just polity needs to be brought about by a particular kind of disposition to have intrinsic normative value. This sense of civic duty is congruent with how people are encouraged to understand their role as citizens in a well-functioning democratic polity. ¹⁷ A democratic and just polity cannot function effectively and be stable over time in the absence of politically committed citizens who are willing to observe laws that serve the common good.

It is then the existence of a particular disposition among partisans that turns partisanship into a vehicle that expresses a commitment to promoting the public good. Good government, understood as a government that serves the public good, is a ‘rich’ or ‘thick’ good that requires citizens to provide each other with the corresponding thin benefit of civic commitment (see e.g. Pettit, 2015). Partisanship has intrinsic normative value when it takes the form of a civic commitment to a particular conception of the public good which is expressed and supported by political parties. Defenders of partisanship (White and Ypi, 2016) are right to point out that party structures can act as catalysts for the provision of the ‘thin’ good of civic virtue that is necessary for the ‘thick’ good of good government, due to the pivotal institutional role that they have in politics and government. But partisanship does not necessarily and robustly serve that role in the absence of the right moral disposition. Without such a disposition, partisanship lacks the intrinsic qualities which characterise a political commitment to the ideals of democracy and justice, and its relationship to them becomes contingent and fragile. Partisanship is often the outcome of such civic commitment, and it often facilitates its expression, but it is not synonymous with such motivation. When it lacks this disposition, it also lacks intrinsic normative value (Muirhead, 2006, 2011; White and Ypi, 2010, 2011).

Partisanship as Contestation

As stated, there are two distinct normative defences of partisanship: one considers partisanship as having intrinsic moral value and the other treats partisanship as having instrumental value. Until now, I have focused on the intrinsic defence of partisanship; I now turn to the instrumental defence. However, considering partisanship instrumentally valuable is not a defence of partisanship as such, but a qualified defence of its benign implications for ideals such as democracy and justice. The position that I defend here is that the contestation characteristic of partisanship matters for good government, even if it does not directly express a commitment to democracy or justice. In what follows, I defend my argument that contestation among partisans is instrumentally good for the normative ideals of democracy and justice, and argue that a world containing no contestation of partisan commitments would be in one respect worse (for both partisans and non-partisans) than one where there is such contestation. I argue that partisans are not necessarily against contestation of their views, as some opponents of partisanship have suggested (Mansbridge et al., 2010). Furthermore, I claim that viewing contestation as necessary for democracy and justice does not entail or require considering partisanship as synonymous to partiality. A belief in the value of partisan contestation is also consistent with treating one’s principles as true and impartial.

Those concerned with the deliberative aspect of democratic politics point out that strong identification with a party is often motivated by devotion to a cause and antagonistic towards the idea of impartial contestation (Guttmann and Thompson, 1996; Johnson, 2006). In other words, they assert that debate among partisans is most often motivated by the desire to rectify injustice in the name of a noble cause, and that contestation only occurs because the partisan’s opponents are seen as flawed. In this view, partisans would be happy to live in a world in which their opponents have conceded that their views are flawed; a world indeed in which the partisan’s views would no longer be contested. Hence, a partisan would not consider political contestation to have value as a form of deliberation.

There are two ways to proceed for those who wish to defend partisanship from this view or criticism. The first is to concede that partisans must accept that contestation is necessary for the ideals of democracy and justice to be both achieved and maintained. But some quickly add that for partisans to accept the importance of contestation, they must necessarily come to see their political judgements as partial (Muirhead, 2006: 721; Rosenblum, 2008): they must see that political judgements involve reconciliation of ‘fractional truths’ to which we are often accidentally committed. I reject this response. However, I also reject what might appear to be a second way to proceed: the claim that there is no value in contestation as such, but only in those forms of partisan contestation which support one’s cherished conception of democracy and justice. The reason I reject this second response is because, unlike the proponents of the first view, I think that an appeal to both truth and contestation is compatible. I support the idea that a world in which even certain truths are contested is better for all (partisans included) than one where there is no such contestation. I hope to show that partisans do not need to consider their views to be partial in order to value the contestation typical in party politics. They have good reasons, I believe, to consider contestation as necessary to democracy and justice even if they think their views are true. I now turn to Mill’s On Liberty, since the defences of partisanship above draw support from this text.

The starting point for those wanting to argue that partisan contestation necessarily entails partiality is Mill’s On Liberty. The following passage is often quoted:

Truth, in the great practical concerns of life is so much a question of reconciling and combining of opposites, that very few have minds sufficiently capacious and impartial to make the adjustment with an approach to correctness, and it has to be made by the rough process of a struggle between combatants fighting under hostile banners (Mill, 1989 [1859]: 48–49).

Later in the same paragraph, Mill also argues that:

On any of the great open questions just enumerated, if either of the two opinions has a better claim than the other, not merely to be tolerated, but to be encouraged and countenanced, it is the one which happens at the particular time and place to be in a minority. That is the opinion which, for the time being, represents the neglected interests, the side of human well-being which is in danger of obtaining less than its share (Mill, 1989 [1859]: 49; emphasis added).

There is a tension in Mill’s writings which is evident in the two excerpts above. In the first excerpt, he advocates a politics of contestation because ‘truth’ is ‘so much a question of reconciling and combining of opposites’. Nonetheless, Mill is concerned with ‘minds sufficiently capacious and impartial to make the adjustment with an approach to correctness’. In the second excerpt, Mill seems to be worried about the ability of those in the minority to obtain what Mill considers to be their objectively fair share. At stake in both texts is a concern for contestation. Mill believes that a greater harm is done by suppressing a view, even if that view is false, than by allowing it to be freely expressed and therefore making it available to public contestation. Mill is ultimately interested in testing what we take to be true by contesting it. If the contested view is in some way true, then suppressing contestation denies us the opportunity to understand the view’s full value.

If the view is false, then suppressing it before subjecting it to public debate deprives us of the opportunity to publicly refute it and hence of recognising truth via ‘its collision with error’ (Mill, 1989 [1859]: 20, 37). This form of instrumental defence allows for the perfectly consistent conclusion that a partisan who is strongly committed to her convictions has at least one good reason to want a plurality of partisan views to be expressed. The public contestation of views, as well as of partisan opponents’ views, preserves all views, including hers, from becoming ‘dead dogma’.

Furthermore, an appeal to truth is central to one’s convictions. The fact of another’s strong identification with party X or with view Y is not reason enough for me to abandon my contestation of party X or view Y. Another’s desire to express partisanship by contesting me does not mean that such partisanship necessarily encapsulates a truth or that I should consider that view valid. Defending partisanship need not depend on any particular metaphysics of truth. Rather, it should rely on the fact that commitment to a party and its ideals may fuel the contestation necessary for us to be reminded of the soundness (as well as the loopholes) of our reasoning. The fact that we can allow for the possibility that our reasons may be contested does not mean that our reasons are flawed. They might be flawless, in fact so flawless as to be constantly alerting our opponents to their reasons’ flaws. Hence, partisanship does not necessarily entail and entrench partiality.

Calling for the abandonment of an appeal to truth in partisan contestation is also problematic in another respect. ¹⁸ The dissociation of partisan contestation from an appeal to truth deprives the very act of contestation of an important motive. Treating one’s commitments as partial diminishes the motivation to contest others’ commitments on the basis of one’s convictions. One believes one’s commitments to be important because one takes them as true in some respect. If one were to take them as partial, then they would not be as important (Christiano, 2008: 214, 219–223).

The danger here is that an appeal to truth can cloud partisans’ judgement. The fact that I contest my opponents and they contest me is no guarantee that my partisan commitments will weaken when they face the force of not just irrational but also rational argumentation. ¹⁹ But perhaps this is the best or second best that we can hope for in the non-ideal world of party politics and the best means available to approximate ideals of deliberative democracy and social justice in these circumstances. Partisan contestation surely forces partisans to come up with supporting reasons for their commitments. This might do little to convince their hardened partisan opponents but it could lead to a state of ‘internal’ impartiality among co-partisans in their quest to provide such reasons. Partisanship contestation between partisans also provides more information to independents, who likewise have access to such debates in the public sphere. In that respect, we could speak of an informational spillover effect of partisan contestation. ²⁰ Partisan contestation, in that sense, encourages the exchange and provision of information between co-partisans and non-partisans, even when it does little to convince partisan opponents. However, thinking that political judgements can be fully articulated and agreed upon without the helping hand of partisan commitments is to deny the reality, and centrality, of party politics in our current, and far from ideal circumstances. We need partisanship and partisans, instrumentally speaking, because they are most likely to fuel electoral campaigns, rally support for party causes and participate in protests and activist actions, as these are the only means available to provide citizens with the motivation and information they need to strengthen democracy and justice (Finkel and Opp, 1991; Mair, 2013; Young, 2001). ²¹

There is potentially a tension here between the intrinsic defence presented in the previous section and the instrumental defence presented in this section. If partisan contestation produces a public benefit, as the instrumental defence suggests, then maybe it does so precisely because of individual self-interest expressed in the form of partisan commitment to a political party. ²² The objection is as follows: emphasising the importance of the structure of partisans’ moral motivation – as the intrinsic defence does – is irrelevant, if not a hindrance, to bringing about what has value: the contestation necessary for good government. Partisan self-interest is sufficient in achieving that. ²³ The two defences of partisanship, the intrinsic and the instrumental, could be treated as separate and self-standing, but I believe they are linked. Two issues are crucial here. First, is the public benefit brought about by partisanship only producible by self-interest? Second, does the intrinsic defence, as a moral concern for the public good, add (anything of) value to civic contestation? I argue that in those cases involving civic commitment, the value of partisanship becomes more robust. My argument, in short, is that an attitude or motivation for acting for the public good serves to achieve, and better guarantees, a government that serves the ideal of democracy and justice.

Imagine a situation where a group of partisans present their self-interest as serving the public good, say by arguing that a tax cut to their socioeconomic group would be good for the economy. One could argue that in those circumstances something like a ‘civilizing force of hypocrisy’ is at play. ²⁴ Partisans have to disguise their self-interest in the language of the public good to justify their commitment to a tax policy which favours them and their party followers. A concern for the public good is clearly instrumental here, and it also serves as a filter for some views that are motivated by brute self-interest, which may only serve factional interests in its absence. Partisans cannot say publicly that they want a tax cut simply because it serves their self-interest, and this shows something important about the regulative effect that a commitment to the common good can have on partisan contestation. ²⁵ The civic norm defended in the section ‘Partisanship as Commitment’ is more comprehensive and demanding than a merely regulating norm operating under the auspices of the civilising force of hypocrisy. But the force of the argument presented here goes further and supports that position. Partisans who are genuinely motivated by a concern for the common good do not need to present their self-interest as serving the public interest, and hence they are more likely to support policies which promote it even when such policies do not necessarily favour them. Hence, the likelihood that partisans will espouse policies which serve factional interests, presented as serving the public interest but in reality undermining it, will be smaller in those circumstances. There will be a smaller number of cases of ‘false positives’ of the ‘civilising force of hypocrisy’, so to speak, when a significant number of partisans are to some extent, or fully, motivated by a concern for the public good. Hence, the moral direction of the first defence secures better outcomes for partisan contestation. ²⁶

Conclusion

As noted at the outset, a normative appraisal of the value of partisanship cannot but be nuanced. This article has examined some recent normative defences of partisanship and supported the claim that such defences can be intrinsic and instrumental in character. After critically assessing the claim that partisanship is intrinsically linked to normative ideals and finding it lacking, I defended the view that what is valuable in partisanship is that it could serve not as a mere reflection of party fidelity but also as an institutional incubator for a civic ethos of political commitment. I then contested the view that an instrumental defence of partisanship’s role in promoting political contestation requires seeing political judgements as partial, and showed that partisan contestation and an appeal to truth are both compatible and beneficial for achieving the ideals of democracy and justice.