Can Truth ( or Problem-Solving ) Do More for Democracy ?

This essay is part of a dossier on Cristina Lafont’s book Democracy without Shortcuts.


Can Truth (or Problem-Solving) Do More for Democracy? Justo Serrano Zamora
In the following comments I discuss Cristina Lafont's critique of democratic epistocracy formulated in her impressive book Democracy without Shortcuts. In particular I address her rejection of the idea that the notion of truth (or of good outcomes) should play any role in justifying the scope and the procedures of democratic decision-making. 1 Rather, we should ground our normative reflections on democracy merely on intrinsic grounds; more concretely, on the normative ideal of self-determination. This rejection, then, not only concerns pure epistemic justifications of democracy, but also mixed models combining intrinsic and epistemic justifications. The latter is represented by Hélène Landemore's approach in her book Democratic Reason (2013) as well as in more recent articles (Landemore 2014a). According to Lafont's reading, in Landemore's mixed model an intrinsic justification of the scope and procedures of democratic decision-making should have normative priority over epistemic justifications (Landemore 2013, 90). However, the latter should also have a normative weight to the extent that they are able to "prescribe maximal scope and thus favor total inclusion in political decisionmaking" (Lafont 2020, 96). Lafont rejects Landemore's mixed model for two main reasons.
Firstly, because epistemic arguments are unable to set a limit for the demarcation of the political community, being in contradiction with higher value intrinsic arguments which make the former arguments futile. Secondly, because although Landemore has been able to present convincing epistemic arguments for maximalizing democratic inclusion, these arguments still fail to account for those aspects such as mutual justification that are essential to the ideal of political self-determination.
The aim of my comments is to defend the possibility of an alternative mixed model, one that, similar to Landemore's, does not give up on the idea that reaching good outcomes should play a normative role in justifying democracy. This mixed model also prioritizes intrinsic reasons for determining democratic procedures and inclusion. At the epistemic core of the model lies John Dewey's notion of inquiry, understood as a method for identifying, defining, and effectively solving collective problems. By relying on Dewey's notion of inquiry my mixed model becomes less vulnerable to Lafont's criticisms. 2 This should be a reason for not rejecting the possibility of a mixed model that includes an outcome-based justification of democratic inclusion and procedures. My argument has three parts: first of all, I present Lafont's criticisms to Landemore's mixed model, focusing on its epistemic dimension. Secondly, I introduce Dewey's notion of experimental social inquiry as an alternative to Landemore's epistemic justification of democracy and show how the former is able to resist Lafont's criticisms. Thirdly, I briefly (re-)consider in what sense a mixed model based on Dewey's ideas is compatible with (and can positively contribute to) Lafont's view of a democracy without shortcuts.

Lafont's arguments against democratic epistocracy (and against a mixed justification of democracy)
According to Lafont, Hèlene Landemore has formulated a compelling epistemic, outcomebased case for democracy by drawing on the idea of collective intelligence. In a nutshell, Landemore argues that the outcomes of a political decision-making process will be truer or better if enhanced by the collective intelligence that merges the internal diversity of perspectives, points of view, and information of all citizens. Cognitive diversity, rather than the aggregation of individual competences, is responsible for the enhancement of the quality of the outcomes of a political decision-making process. This represents an epistemic argument for democracy since the more inclusive a group is, the more diverse it is 3 and the better the quality of political outcomes which can be expected. Hence, Landemore's argument rules out the possibility of justifying the government of a minority of experts on epistemic reasons. This is so because the sum of their individual competences cannot make up for the epistemic benefits of cognitive diversity. As mentioned in the introduction, Landemore believes that this epistemic justification can be introduced in a mixed model of justification. According to this model, an intrinsic justification should have priority in determining the demarcation of political inclusion and the democratic procedures of decision-making. However, epistemic arguments should also play an independent role in maximalizing the range of those who can participate in decision-making.
Lafont formulates four main critiques to Landemore's mixed model, the first of which refers to the possibility of providing a mixed model in the sense just mentioned. According to Lafont, while Landemore is able to justify maximal inclusion on the basis of cognitive diversity, her epistemic argument is not intrinsically capable of setting a clear line of demarcation between those who can legitimately participate in decision-making and those who cannot. Indeed, why not include members of other political communities in decision-making if they have more information or simply know how better to solve a certain collective problem? For Lafont, the very impossibility of setting a line of demarcation capable of overlapping with the set of those who can legitimately participate in political decision-making weakens the epistemic strategy of justification; therefore, the mixed model should be rejected. However, it is not clear why this should be the case. Given the fact that in Landemore's model the epistemic justification of democracy is subordinated to the intrinsic one, it seems reasonable for this model that certain citizens may be excluded from the political process on intrinsic grounds. This does not make the epistemic justification of democracy merely superficial. Citizens not only praise democratic decision-making for its capacity to embody values such as self-determination and equality, but also for its capacity to effectively resolve problems. So why not strengthen the motivational basis for democracy, provided by the intrinsic argument, with a complementary outcome-based justification?
The rest of Lafont's critiques challenge the democratic consistency of Landemore's own epistemic justification. Here I will briefly present them, showing in the next section how Dewey provides a better prepared alternative. The second criticism concerns a further element of Landemore's epistemic view that has yet to be mentioned. Landemore is aware that in actual democracies there is too great a number of citizens for direct decision-making participation, making the existence of representative institutions necessary. In order to deal with this problem she proposes to introduce groups elected by lottery which fulfil the condition of cognitive diversity. According to Landemore these groups are sufficiently justified, from an epistemic point of view, to take decisions for the whole political community. Lafont considers this lottocratic solution to the problem of representation to be a good example of the kind of democratic shortcut she is opposing, because this solution forces the majority of citizens to blindly defer to the deliberations of the lottery-selected group, and hence makes probable the permanent misalignment of their preferences with the outcomes of the political process. For Lafont, it is precisely this possibility of misalignment which undermines the principle of self-determination that lies at the core of the democratic ideal. These kinds of lottocratic solutions to the problem of representation -even if they are epistemically justified -should therefore be rejected since they represent institutional shortcuts that undermine the ideal of democracy.
To this criticism Lafont adds a further consideration regarding the very plausibility of the epistemic argument Landemore provides. According to Lafont, Landemore's argument focuses too much on tracking truth and leaves aside another fundamental aspect of democracy's epistemic dimension, namely the "justifiability of the policies in question to those who must comply with them." (Lafont 2020, 98, author's emphasis). For Lafont, guaranteeing the mutual justifiability of policies -ensuring that they can be accepted as reasonable by citizens in a process of mutual give and take -represents a requirement for citizens actually endorsing and identifying with them. This clearly affects Landemore's lottocratic proposal, since it excludes a large group of citizens from the mutual practices of justification which are essential to any democratic decision-making process.
Finally, it is precisely the dismissal of mutual justifiability in Landemore's epistemic proposal that is the source of a further deficit in her view, namely, the irenic understanding of democratic political processes. According to this latter view, even if disagreement is certainly present in political debate, Landemore assumes that "once decision-makers hit on the right political answers, agreement by decision-takers will simply follow." (Lafont 2020, 99). This view is problematic, firstly to the extent that it dismisses mutual justification as the condition for solving disagreements, which, as we saw, represents a quintessential element of democratic decision-making. But it is also the reason behind the dismissal of conflict and struggle which 'deep democrats' reproach 'epistemic democrats' with. By assuming that conflict can be easily overcome by somehow pointing to true or better outcomes, Landemore and other epistemic democrats eradicate the necessary conflicts that arise in processes of mutual justification from the background of existing disagreements. These conflicts are an essential part of the political process in plural societies, providing the condition for collective learning processes by which final agreements on better outcomes can be reached.

John Dewey's notion of Experimental Social Inquiry (ESI)
In recent decades, John Dewey's work has become a valuable source for those seeking to provide democracy with an epistemic footing (Cf. Anderson 2006 andBernstein 2010). In particular, Dewey's experimentalism has been mobilized in accounts of democratic institutions, public deliberation, and social movements (Cf, for example, Sabel 2012, Ansell 2011, Frega 2019 and Serrano Zamora 2017). In this section I will present Dewey's notion of experimental social inquiry 4 (ESI), focusing on those elements that are relevant for engaging with Lafont's criticisms. Indeed, I aim to show that a Dewey-based epistemic justification of democracy: 1) does take mutual justifiability as a necessary part of the problem-solving process; 2) is able to avoid the lottocratic shortcut proposed by Landemore; and 3) takes the idea of conflict as an inherent part of the inquiry-process in cases of existing disagreements. In the third and final section I will briefly focus on what it can mean to mix Dewey's epistemic arguments with an intrinsic justification for democracy, and how this strategy can reinforce Lafont's own participatorydeliberative approach.
For Dewey, ESI does not primarily consist in finding or discovering a pre-existing social truth.
Rather, through ESI citizens identify and articulate social problems, propose solutions, implement them, and test their capacity to solve those problems (and revise them if they are not successful). ESI is not only about solving problems that are already given in a completed, fully determinate form, it is also about defining and solving problems in a way that they are gradually determined as certain problems. Hence, as Rahel Jaeggi puts it, for Dewey social problems are both "given and made," "subjectively constructed and objectively given." (Jaeggi 2018, 140). On the objective side, problems are given crises affecting our practices or interpretations.
However, the very concrete form of a problem depends on the process of interpretation of the inquiry leading to the overcoming of the crisis. Jaeggi understands this process as a "gradual hermeneutic adjustment of the problem and its description." (Jaeggi 2018, 144).
The characterization of inquiry as a progressive determination of problems is at the heart of Dewey's experimentalist understanding of ESI. According to this, the very operations previously mentioned as being part of ESI (identification, articulation of problems, generation of solutions as hypothesis, implementation, test and revision) are mutually dependent. So, for example, in an experimental account, we cannot first provide a definitive definition of a problem and then figure out its solution. The process by which we define problems must be linked to the processes through which we look for solutions; problem-definitions may therefore vary This very brief characterization should suffice to show that Dewey's model does not fall prey to the three above critiques -and to point to a further advantage of this view that will be developed in the last section. Regarding her first criticism: the need for clarification regarding the difference between the notion of truth at the core of Lafont's remarks and Dewey's notion of inquiry as problem-solving; while for Lafont a truth or outcome-based epistemic justification would set the scope and procedures that are necessary to figure out the right outcome of a decision-making process, a Deweyan account puts emphasis in actually solving a problem.
Surely, in order to solve the problem we need to find the right or best solution (the truth 6 ) to a problematic situation, but the solution does not become a solution to a problem until it has been fully implemented and tested. This has consequences for the present discussion since it puts into question Lafont's strong differentiation between figuring out the truth and mutually justifying our views and positions within the decision-making process. Hence, if there is no way of figuring out the best outcome of a political process than by implementing and testing it, mutual justification cannot be other than a necessary aspect of effectively solving a problem.
Indeed, without mutual justification we could not reach the agreements that are necessary for implementing a solution, nor could there be any stable solution to a social problem (Dewey 1938, 465).
Secondly, Dewey's notion of ESI can hardly be compatible with a lottocratic shortcut of the kind Landemore proposes. Hence, the very idea that ESI is an experimental process, in the previously stated sense, makes citizen participation necessary in all stages of inquiry. Again, this involves the process of implementation and testing of solutions too. Dewey's epistemic justification does not allow this kind of democratic shortcut, not only because mutual justifiability is a necessary part of the process of resolving a social problem, but also in the sense that the problem is not fully determinated and solved until it has been tested by those affected by it.
Finally, an ESI-based justification of democracy does not reject the need for political struggle for overcoming existing disagreements as part of the experimental process. Dewey himself combines an epistemic approach to democracy with a theory of social struggle wherein oppressed minorities struggle in public life for the recognition of their rights. In this sense, the need to reach better outcomes in politics involves collective organization with the aim of developing and struggling for the visibilization of facts and interpretation of norms which are alternative to the hegemonic views in a society. Thinking of politics as a collective process of inquiry involves both thinking of politics as a sphere of epistemic cooperation and also as a conflict-ridden sphere where what is taken to be the "truth" is put into question by those who have been excluded from the inquiry process. By introducing the notion of ESI and understanding collective actors as inquirers, struggle can be thought of as a collective learning process where disagreements can be potentially overcome if it is carried out in reasonable ways which promote collective learning processes. 7

Conclusions: in the search for a mixed model
As I have shown, the Dewey-based epistemic justification of democracy is maximally inclusive, takes mutual justification as an essential part of the problem-solving process, does not allow any lottocratic shortcuts, and takes disagreement and conflict as essential elements of the political process. Is it possible to provide a mixed model for justifying democracy in light of these reflections? I have previously characterized this model as one of subordination: once the group of citizens who can legitimately participate in political decision-making has been defined by the actual interpretation of this principle is set by epistemic criteria. There are good reasons to defend such a model. Hence, even if it might collide with intrinsic justifications, the epistemic justification can play a valuable (additional) motivational role in promoting participation and deliberation. Furthermore, one might want to consider if Dewey's idea that solutions need to be tested and revised may be able to provide an epistemic underpinning to the epistemic legitimacy of judicial review. But there is another consideration that is worth mentioning here, and which makes the prospects of a mixed model particularly interesting. It can be formulated as follows: if the epistemic justification of democracy was able to propose new forms concerning scope and procedures that go beyond established understandings of democracy, why not consider them as a valuable source for democratic innovation? According to the hierarchy of values in a mixed model, epistemically motivated innovations -i.e. innovations that are due to citizen's efforts to identify, define, and resolve problems -are not by themselves valuable from a democratic point of view. In order to be valuable, and this point is crucial, they need to be appropriated as better realizations of intrinsic ideals, such as Lafont's idea of self-determination. However, as far as epistemically motivated practical innovations could be seen as realizing self-determination in different (or even better) ways, we could identify in epistemic justifications a productive source of democratic innovation and of the expansion of our political imagination. 8 Be that as it may, Lafont does not need to accept this last consideration in order to be ready to endorse a Dewey-based mixed model that has the idea of problem-solving at the core of its epistemic aspect. Hence, by showing that effectively solving a problem is different to figuring out the truth, I hope to have shown that a Deweyan, ESI-based model is able to resist Lafont's concerns regarding the plausibility of an outcome-based justification of democracy. For such an experimentalist model, democratic shortcuts to defining and solving our collective problems are not possible.
Notes 1] Lafont agrees with epistemic democrats that the idea of truth or better outcomes can contribute in justifying democracy against sceptics. This should not be confused with the role it can play in determining the scope and the procedures of democratic decision-making, which Lafont denies. Here I will use the expression "justifying democracy" always in this second sense. 2] Note that through the text I am assuming that Lafont's interpretation of Landemore's arguments is right. 3] Landemore (2013, 233) adds that this is only the case for liberal societies. 4] By social experimental inquiry, Dewey refers to inquiry on social and political matters (Dewey 1938). Social inquiries include sociological inquiries and journalistic research but also, most importantly, political decision-making to the extent that it involves the identification, definition and solving of collective problems (see Dewey 1927). The meaning of experimental inquiry is spelled out in this section. 5] Though it cannot here be further expanded, I believe that the idea that problems are to a certain degree indeterminate can also be operationalized for a justification of maximal inclusion. This argument would come close to Landemore's own strategy consisting in justifying maximal inclusion and political equality on the basis of "radical uncertainty" of social problems (Landemore 2014b). 6] Dewey (1938, 11) himself uses the expression "warranted assertions". 7] See also Anderson (2006) for a discussion on Dewey's epistemic approach to democracy and the role of political disagreement and dissent. 8] I defend this idea concerning social movements in Serrano Zamora (2017).