The Methodologically Flawed Discussion about Deep Disagreement

Abstract

Questions surrounding deep disagreement have gained significant attention in recent years. One of the central debates is metaphysical, focusing on the features that make a disagreement deep. Proposals for what makes disagreements deep include theories about hinge propositions and first epistemic principles. In this paper, I criticize this metaphysical discussion by arguing that it is methodologically flawed. Deep disagreement is a technical or semi-technical term, but the metaphysical discussion mistakenly treats it as a common-sense concept to be analyzed and captured by our pre-theoretical intuitions. Since the literature on deep disagreement is subject to this fundamental confusion and deep disagreement is not a helpful umbrella term either, I propose eliminating the notion of deep disagreement from the philosophical discourse. Instead of analyzing the nature of deep disagreement, we should develop theories about different forms of disagreement, including disagreement about hinge propositions and disagreement about epistemic principles, and, in particular, a theory of rationally irresolvable disagreement.

1. Overview

Deep disagreement is broadly understood as a disagreement between two or more parties that is in some sense systematic, persistent, and hard to resolve. Different aspects of deep disagreement are discussed, for instance whether it is rationally resolvable or whether it has some impact on epistemic relativism.¹ One central project concerning deep disagreement is the metaphysical program of inquiring into its metaphysical nature, i.e., what deep disagreement is and in what sense it is deep. In this paper, I will argue that this metaphysical program is doomed to fail because in this discussion deep disagreement is analyzed like a common-sense notion such as knowledge, belief, or truth. Deep disagreement, however, is a technical or semi-technical term, introduced by Fogelin with a stipulative definition, which need not accommodate any pre-theoretical intuitions about the target notion. Consequently, adjusting theories about deep disagreement to intuitions, as we do when analyzing concepts such as knowledge, belief, or truth, is mistaken.

In the face of this methodological flaw, we should only continue theorizing about the notion of deep disagreement if doing so is somehow pragmatically useful. However, I will argue that there are no practical advantages to continuing to think about deep disagreement. The current discussion about deep disagreement reveals misunderstandings and confusions about deep disagreement among various dimensions, namely concerning the nature of deep disagreement, about paradigmatic cases of deep disagreement and about its rational resolvability. I conclude that the notion of deep disagreement should be eliminated from current discussions about disagreement. We should rather distinguish different forms of disagreement involving disagreements that are based on hinge propositions or fundamental epistemic principles and, in particular, we should theorize about the various forms of rationally irresolvable disagreement.

I proceed as follows: In section 2, I summarize Fogelin's (1985) pioneering work on deep disagreement, including how he introduced the stipulative notion. In section 3, I present the current debate about deep disagreement, including its central questions and the main schools of thought about the nature of deep disagreement, theories based around hinge propositions, and others centered on first epistemic principles. In section 4, I then analyze the methodology of existing approaches to deep disagreement, exploring in particular the metaphysical program that aims at investigating the nature of deep disagreement via orthodox philosophical methodology involving concept analysis, descriptive definitions, or explanations. In section 5, I argue that this metaphysical program is methodologically flawed since it takes deep disagreement to be a common-sense notion to be analyzed, such as knowledge, belief, or truth. In section 6, I consider an alternative interpretation of the current debate and argue that this approach faces similar problems. In section 7, I then investigate the remaining options given this methodological defect. I first reflect on the possibility of preserving the notion of deep disagreement despite these difficulties, ultimately finding this option wanting since using this notion causes more confusion than clarity. I then conclude that we should abandon the notion of deep disagreement and focus instead on various kinds of systematic disagreement, and in particular, focus on developing a theory of rationally irresolvable disagreement.

2. Fogelin on deep disagreement

Fogelin (1985) introduces the notion of deep disagreement in his pioneering paper “The Logic of Deep Disagreements.” He begins by pointing out that, in normal contexts, two parties to an argument interact based on commonly shared assumptions, but in non-normal contexts, they lack this common ground. Fogelin then focuses on the specific features of disagreements in such non-normal contexts. In the following passage Fogelin (1985: 7f) introduces the notion of deep disagreement:

What happens to arguments when the context is neither normal nor nearly normal? The answer that seems forced upon us is this: to the extent that the argumentative context becomes less normal, argument, to that extent, become impossible. This is not the […] weak claim that in such contexts arguments cannot be settled. It is the stronger claim that the conditions for argument do not exist. The language of argument may persist, but it becomes pointless since it makes an appeal to something that does not exist: a shared background of beliefs and preferences. Here I wish to speak about deep disagreements. My thesis, or rather Wittgenstein's thesis, is that deep disagreements cannot be resolved through the use of argument, for they undercut the conditions essential to arguing. (Fogelin 1985: 7f)²

After introducing the notion of deep disagreement, Fogelin then proceeds to distinguish between different sorts of disagreements. First, he demarcates deep disagreement from other forms of disagreement:

What is a deep disagreement? First let me say what I don't mean by this notion. A disagreement can be intense without being deep. A disagreement can also be unresolvable without being deep. I can argue myself blue in the face trying to convince you of something without succeeding. The explanation might be that one of us is dense or pig-headed. And this is a matter that could be established beyond doubt to, say, an impartial spectator. But we get a very different sort of disagreement when it proceeds from a clash in underlying principles. Under these circumstances, the parties may be unbiased, free of prejudice, consistent, coherent, precise and rigorous, yet still disagree. And disagree profoundly, not just marginally. Now when I speak about underlying principles, I am thinking about what others (Putnam) have called framework propositions or what Wittgenstein was inclined to call rules. We get a deep disagreement when the argument is generated by a clash of framework propositions. (Fogelin 1985: 8)

In this passage, Fogelin points out that deep disagreement differs from other forms of disagreement that are hard or impossible to resolve due to obvious epistemic vices such as stubbornness. He then characterizes deep disagreement by saying that it results from a “clash in underlying principles” diagnosing deep disagreement as having a holistic character beyond disagreements that only concern single propositions:

[W]hen we inquire into the source of a deep disagreement, we do not simply find isolated propositions […], but instead a whole system of mutually supporting propositions (and paradigms, models, styles of acting and thinking) that constitute, if I may use the phrase, a form of life. (Fogelin 1985: 9)

Fogelin finishes his paper by expressing his pessimism about the rational resolvability of deep disagreement when noting:

In the end, however, we should tell the truth: there are disagreements, sometimes on important issues, which by their nature, are not subject to rational resolution. (Fogelin 1985: 11)

So far, this is Fogelin's take on deep disagreement. His paper addresses a topic that most readers can intuitively grasp. However, in other ways, Fogelin's presentation is rather unclear, e.g., about the nature of deep disagreement or about typical examples of deep disagreement. Consequently, his theory about deep disagreement has provoked numerous replies, objections, and suggestions for improvement.

3. The contemporary discussion about deep disagreement

If we consider the notion of deep disagreement without any philosophical background, then the notion seems ambiguous between referring to a disagreement that is deep in the sense that it is persistent and hard to resolve and to a disagreement about deep, perhaps philosophical, issues. While disagreements about deep questions are often hard to resolve, there is nevertheless a clear distinction between these two ways of understanding deep disagreement. Unsurprisingly, the existing literature seems to shift between these two versions of deep disagreement. Fogelin's (1985) paper mirrors this ambiguity when he characterizes deep disagreement as fundamental and irresolvable but presents as examples for deep disagreement philosophical disputes about abortion and the fairness of affirmative action, which are intuitively disagreements about deep questions rather than rationally irresolvable disagreements. Other authors also shift between these potential meanings of deep disagreement. Lynch (2010) focuses on the rational irresolvability of deep disagreement, while Feldman (2005) assumes that deep disagreements are about deep philosophical questions and stresses their rational resolvability via suspension of judgment.

Contemporary theories of deep disagreement currently center around the following two questions:

Q1: What is deep disagreement?

Q2: Is deep disagreement rationally resolvable?

Q1 addresses the metaphysical nature of deep disagreement. Concerning this question, two major schools have emerged. One account holds that deep disagreement is disagreement about hinge propositions or hinge commitments, while the other view takes it that deep disagreement concerns fundamental epistemic principles. Hinge theories about deep disagreement rely on Wittgenstein's (1969) view that rational evaluation can only take place on the basis of fundamental shared assumptions, or hinges, which remain outside the context of ordinary rational evaluation. Even though he doesn't explicitly mention hinges, Fogelin (1985) invokes Wittgenstein when motivating the notion of deep disagreement. To be attracted to this view, one must accept some form of hinge epistemology.³ But even if one generally buys into hinge epistemology, various questions must be settled on the way to a hinge theory of deep disagreement. The first and most obvious question concerns the content of hinge epistemology, i.e., which propositions, if any, are potential hinges. Wittgenstein (1969) lists as examples propositions about very fundamental facts that one has hands, that one is speaking one's native language, that one's name is such-and-such, that one has never been to the moon, that that there is a brain inside one's skull, that the table is still there when no one sees it, and that the Earth is round. Brueckner (2007: 285) gives examples of hinge propositions that also include epistemic principles and anti-skeptical claims, propositions like that there is an external world, that sense perception is reliable, that I am not a brain in a vat, that my faculty of reasoning is reliable, that the Earth is more than three minutes old, that testimony is reliable, and that memory is reliable. Pritchard (2021: 1120) argues that hinge propositions are hierarchically structured. There is a central overarching “über hinge commitment,” that we are not radically and fundamentally in error in our beliefs, and all other hinge commitments are manifestations of this über hinge commitment.⁴

Along with giving different lists of hinge propositions, hinge theorists also disagree about whether hinge beliefs can be rationally assessed. Epistemic hinge theories answer this question in the affirmative. As a protagonist of epistemic hinge theories, Wright (2004, 2014) argues that we are entitled to trust hinge propositions whereas entitlement is a kind of rational attitude that one can take towards propositions. Non-epistemic hinge theories answer the question in the negative. Pritchard (2016, 2021), for example, argues that hinges cannot be rationally accessed because they cannot be actually believed. Other non-epistemic theories, as defended by Wright (1985) and Moyal-Sharrock (2004, 2016), hold that hinge propositions cannot be rationally accessed because they are not truth apt and, consequently, cannot be true or false.

The other school of thought about the nature of deep disagreement assumes that it is based on disagreement about fundamental epistemic principles. This view is advocated by Lynch (2010, 2016). He argues that fundamental epistemic principles cannot be justified via other principles but only via epistemically circular reasoning, i.e., by reasoning that relies on the correctness of the principle in question. Based on this intuition about fundamental epistemic principles and their epistemically circular justification, Lynch (2010: 264) introduces the notion of deep disagreement as follows.

I'll say a disagreement is deep when it meets the following conditions:

1. 1. Commonality: The parties to the disagreement share a common epistemic goal(s).

2. 2. Competition: If the parties affirm distinct principles with regard to a given domain, those principles (a) pronounce different methods to be the most reliable in a given domain; and (b) these methods are capable of producing incompatible beliefs about that domain.

3. 3. Non-arbitration: There is no further epistemic principle, accepted by both parties, which would settle the disagreement.

4. 4. Mutual circularity: The epistemic principle(s) in question can be justified only by means of an epistemically circular argument.

Since epistemic circular reasoning is not persuasive, deep disagreement about fundamental epistemic principles is not resolvable.

However, also alternative fundamental epistemic principle views are defended. Kappel (2021: 1039) defends a more pluralistic approach toward deep disagreement than Lynch, claiming that in “deep disagreements local disagreements are intertwined with more general basic disagreements about the relevant evidence, standards of argument or proper methods of inquiry in that domain.”⁵

Hinge theories and fundamental epistemic principle theories about deep disagreement face particular challenges and technical problems, but these will not be the focus of this paper.⁶ Rather, I will consider a more fundamental methodological mistake with both of these kinds of views. Before coming to this methodological problem, however, it will be helpful to briefly mention different views about Q2, the question of whether deep disagreements are rationally resolvable.

Fogelin (1985: 9), who introduced the notion of deep disagreement, was explicitly pessimistic about its rational resolvability. Along with introducing the notion of deep disagreement he also asks “But if deep disagreements can arise, what rational procedures can be used for their resolution? The drift of this discussion leads to the answer NONE.” The majority of theorists agree with Fogelin on this point. Lynch (2010: 273), for instance, uses Wittgenstein's metaphor when claiming that when “there is deep epistemic disagreement over some fundamental principle, the disagreement has hit bedrock, the spade has turned,” implying that deep disagreements are fundamentally irresolvable. However, some authors also argue for the rational resolvability of deep disagreement. Lugg (1986) points out that disagreements in general are not only resolved if one of the initially disagreeing parties persuades the other but also when both parties end up suspending judgment about the target proposition. Feldman (2005) takes up this insight and argues that suspension of judgment leading to agreement between both initially disagreeing parties is the rational reaction for many paradigmatic instances of deep disagreement. Matheson (2021) argues that deep disagreement is rationally resolvable since epistemic principles are self-supporting and justified if they meet their own standards of justification. Within his hinge theory, Pritchard (2021) argues that deep disagreement can be resolved in “an indirect, and side-on fashion” if the disagreeing parties look for common ground (common beliefs, common hinges) in order to change their wider set of beliefs. Their über hinge commitment, of course, is ultimately not changeable, but this still allows parties to a deep disagreement could still alter other, less crucial, hinge commitments. However, as Ranalli (2021) and Ranalli and Lagewaard (2022b) note, the notion of rationally resolvability is multiply ambiguous and, consequently, adherents as well as opponents of the rational resolvability of deep disagreement must clarify in which sense deep disagreement is (not) rationally resolvable.

4. The methodology of the contemporary debate

In this section, I will reveal a methodological approach predominant in the discussion about deep disagreement, which can be labeled the metaphysical program. Most authors do not say much about their methodological assumptions and choices. Thus, we have to extract their methodology from the way they proceed. However, in recent papers, Ranalli (2021) and Ranalli and Lagewaard (2022a, 2022b) explicitly reflect on the aims of a theory of deep disagreement and on the appropriate methodology. Ranalli (2021: 983) characterizes the metaphysical program of deep disagreement as follows:

[W]e might think that we should step back and ask about the nature of deep disagreement directly, for example, by asking what such disagreements consist in and what kinds of attitudes are at stake in such cases. As it stands, we lack any clear criteria for what deep disagreement is which is separate from the epistemological question of whether they are rationally intractable or irresolvable.⁷

If we follow this metaphysical project, then we assume that there is deep disagreement and then inquire into its nature, i.e., in what sense it is deep. Thereby, we search for the depth of deep disagreement.

How to analyze the metaphysical program of deep disagreement? One way of interpreting it is in terms of explanations, i.e., as the project of explaining the nature of deep disagreement, which succeeds if deep disagreement and its nature are properly captured. A similar project is the one of analyzing the concept of deep disagreement, a project that typically aims at formulating individually necessary and jointly sufficient conditions for deep disagreement. Various different views on conceptual analysis are taken and it is also disputed whether conceptual analysis is a proper philosophical methodology at all. However, summarizing the program of conceptual analysis, Margolis and Laurence (2022) say that for “proponents of traditional conceptual analysis, the analysis of a concept is successful to the extent that the proposed definition matches people's intuitions about particular cases, including hypothetical cases that figure in crucial thought experiments.”⁸

Another way of interpreting the metaphysical program is in terms of definitions, where providing a definition is related to conceptual analysis. With definitions, we seek individually necessary and jointly sufficient conditions of the target term. Despite the same formal structure as conceptual analysis, different types of definitions can be distinguished. Stipulative definitions define the meaning of a term without considering any potential prior meaning. As Gupta (2021) puts it, they involve “no commitment that the assigned meaning agrees with prior uses (if any) of the term.” Descriptive definitions, in contrast, spell out the meaning of a term that already exists in natural language by considering its existing meaning. For example, philosophers often present descriptive definitions of ‘truth’ or ‘know’. Such descriptive definitions have to match, to a certain extent, with the existing usage. Accordingly, a lack of fit with the existing usage is an objection against a descriptive definition, but not against a stipulative definition. Moreover, there are explications in a Carnapian (1956) sense, which involve elements of stipulative and descriptive definitions. Gupta (2021) describes Carnapian explications as follows:

Sometimes a definition is offered neither descriptively nor stipulatively but as, what Rudolf Carnap (1956: §2) called, an explication. An explication aims to respect some central uses of a term but is stipulative on others. The explication may be offered as an absolute improvement of an existing, imperfect concept. Or, it may be offered as a “good thing to mean” by the term in a specific context for a particular purpose.

The context of the definition and the intentions of the author (and perhaps other pragmatic factors) determine of which type a particular definition is.

Conceptual analyses and definitions usually aim at providing individually necessary and jointly sufficient conditions of the target notion. However, a theory can also be illuminating when focusing on a particularly crucial necessary condition. This holds, for example, for Sosa's (1999) theory of safety which stipulates that safety is necessary for knowledge but leaves open whether it is also sufficient.⁹ Moreover, a view can also be explanatorily fruitful by holding that the target notion is not analyzable in other terms, as Williamson (2000) shows in his groundbreaking analysis of knowledge as fundamental. Accordingly, fruitful theories about deep disagreement can also reveal only crucial necessary conditions or argue that deep disagreement is a phenomenon with a specific nature that cannot be further analyzed.

How can Fogelin's original account of deep disagreement be understood? Here is how he introduces the novel concept of deep disagreement:

The language of argument may persist, but it becomes pointless since it makes an appeal to something that does not exist: a shared background of beliefs and preferences. Here I wish to speak about deep disagreements. (Fogelin 1985: 7f)

In this quote, Fogelin first picks out a particular kind of disagreement, disagreements that lack a shared set of background beliefs and preferences. He then goes on to dub such disputes “deep disagreements.” Before this introduction, the notion of deep disagreement did not exist in the philosophical literature. Fogelin's introduction of the term, understood as a definition, is best taken as stipulative. The situation is different for later philosophical work on deep disagreement. Given the existing literature on deep disagreement, Lynch's conception of deep disagreement as disagreement about fundamental epistemic principles, for example, could also be understood as a descriptive definition or as an explanation. Analogously, hinge theories about disagreement could be understood as providing a descriptive definition or a Carnapian explanation of deep disagreement.

With all of this in mind, how should we approach questions Q1 and Q2 about deep disagreement? Here, at least three different takes are possible. First, we could hold that a theory about the nature of deep disagreement tells us something about its rational resolvability or irresolvability. Accordingly, the metaphysical project of investigating the nature of deep disagreement must also address whether deep disagreement is rationally resolvable. Thus, an answer to Q1 about the nature of deep disagreement also contains an answer to Q2. In this sense, the two questions are not independent of each other. Lynch (2010) takes this option. His criteria for deep disagreement, that the two parties affirm different principles that can produce incompatible beliefs and that there is no further principle accepted by both parties available for settling the question, imply that deep disagreement is irresolvable. Thereby, he incorporates an irresolvability clause in his definition of deep disagreement.

The second option concerning the connection between questions Q1 and Q2 has it that we can develop a theory about whether deep disagreement is resolvable or not without any preliminary theory about the nature of deep disagreement. This path is chosen by Feldman (2005) and Matheson (2021) who argue for the rational resolvability of deep disagreement on the basis of an intuitive understanding of deep disagreement rather than on the basis of a full-fledged theory. However, this methodological approach has its limits. Even if theories focus on the rational resolvability of deep disagreement, they must presuppose some view about deep disagreement in order to make their point. Feldman (2005), for example, assumes in line with Fogelin (1985), but in contrast to current orthodoxy, that deep disagreement is usually about deep philosophical issues. Given this background assumption, Feldman argues that mutual suspension of judgment is the objectively rational resolution. This solution is, however, less plausible if we consider cases that are currently regarded as paradigmatic examples of deep disagreement, such as disagreement between a scientist and a creationist about the age of the Earth.¹⁰ In this case, it seems highly implausible to claim that the disagreement is rationally resolved if the intuitively irrational and the intuitively rational person both suspend judgment.

If we instead opt for a third option, the two questions about deep disagreement, Q1 and Q2, are equally important but independent of each other. We can develop a theory about the nature of deep disagreement, but this theory must not entail the resolvability or irresolvability of deep disagreement, because this question has to be independently addressed. This methodological approach is defended by Ranalli (2021: 985):

Now I think that the rational irresolvability of deep disagreement in any of these senses is not a plausible desideratum for a satisfactory theory of deep disagreement. It's not what we should expect from a satisfactory theory of deep disagreement because it's an open and interesting question whether deep disagreements are rationally resolvable in any of those senses. Presumably, it requires a non-trivial argument to show that they're not resolvable in any of those senses. And there might be such arguments […]. But it shouldn't be built into what we expect from a theory of what deep disagreement is that they are rationally irresolvable.

From this point of view, Ranalli (2021) criticizes Lynch for incorporating into his account the irresolvability of deep disagreement, a question that should be open to further investigation once deep disagreement is better understood. Ranalli's view strikes me as implausible. Ultimately, I will argue for eliminating the notion of deep disagreement, but given the assumption that there is deep disagreement, I do not see why a theory about the nature of deep disagreement should not contain as one of its components a view about its rational resolvability or irresolvability. In the literature, the question of whether deep disagreement is rationally resolvable is one (or the) central question. From that point of view, a theory about the nature of deep disagreement should contain a view about its rational resolvability. Thus, under the assumption that theorizing about deep disagreement is useful, I find Ranalli's position implausible. Nevertheless, I take this view hereinafter as one of the possible options to be considered.¹¹

5. The methodological flaws of the metaphysical project

The metaphysical program is one popular methodological approach to deep disagreement, aiming to investigate the metaphysical nature of deep disagreement. This project proceeds analogously to traditional philosophical investigations into the nature of phenomena and/or concepts such as knowledge, belief, or truth.¹² These traditional projects provide an explanation of the target phenomenon by giving descriptive definitions, Carnapian explanations, or conceptual analyses. We usually proceed then by adjusting the results of the proposed analyses or definitions with our pre-philosophical intuitions about the phenomenon and/or concept. If the definition does not match our intuitions, either because it is intuitively too broad and includes cases that should be ruled out or too narrow and excludes cases that should intuitively be included, then the conceptual analysis or the definition are adapted accordingly. For example, Gettier (1963) famously showed that the traditional conceptual analysis or definition of knowledge as justified, true belief is inadequate, just as fake-barn cases showed the same for Goldman's (1967) causal theory of knowing. In both cases, the definition of knowledge turned out to be too wide to capture our pre-theoretical intuitions, since it allows for knowledge in cases that we intuitively want to exclude. This project of adjusting theories and pre-theoretical intuitions to one another faces well-known problems since, at some points, we have to sacrifice either some of our intuitions for the sake of a systematic theory or a systematic theory for the sake of a purely descriptive account of our (often conflicting) intuitions. The challenge that we face is steering carefully between the Scylla of developing a systematic theory that is not in line with our intuitions and the Charybdis of merely collecting empirical data about intuitions without developing a systematic theory.¹³ Nevertheless, the overall approach is sufficiently clear and well-established philosophical methodology.

At this point, a difference between the analysis of deep disagreement and the analyses of other philosophical concepts such as knowledge, belief, or truth becomes obvious. These are all fundamental common-sense notions and rather stable over natural languages. In conceptually analyzing these notions, we aim to find a systematic theory that matches (as well as possible) our pretheoretical intuitions about these common-sense concepts. If existing theories do not meet our intuitions, as Gettier showed for the JTB view of knowledge, then the theory is abandoned. Deep disagreement is analyzed as if it were such a common-sense concept but, in fact, it is not. Rather, it is a technical or semi-technical notion that was originally introduced by Fogelin via a stipulative definition to describe a particular type of disagreement.¹⁴

If deep disagreement is a purely technical term, then there is no commonly shared concept of deep disagreement and no corresponding common-sense notion in any natural language. We can test a given theory about common-sense concepts against our intuitions by trying to find intuitive counterexamples. We cannot proceed in the same way with technical terms, since there are no pretheoretical intuitions, it claims to be capturing. However, more plausibly, deep disagreement is a semi-technical term in the sense that people can understand both words along with their semantic combination. Nevertheless, different subjects plausibly have different intuitions about the meaning of “deep disagreement,” e.g., whether it is about deep questions or whether the disagreement itself is deep and about the sense in which it is deep. Thus, there is no common intuition about deep disagreement to be captured. Analyzing purely individual intuitions, on the other hand, might be successful insofar as philosophers have such individual intuitions. However, one cannot argue that such a conceptual analysis provides the correct theory about the nature of deep disagreement since the scope of the investigation includes only individual intuitions. In both cases, we are not restricted by any pretheoretical intuitions, neither common nor individual. In this sense, there are no correct or incorrect, no adequate or inadequate theories of deep disagreement. Rather, any theory about a form of disagreement that is deep in some sense could be called a theory of deep disagreement.¹⁵

Furthermore, with common-sense notions, we usually share intuitions about paradigmatic instances of the target notion although there might be different intuitions about cases at the periphery. For example, true beliefs based on perception in ideal circumstances and given all relevant discriminatory capacities constitute paradigmatic cases of knowledge, while true beliefs lacking discriminatory capacities or formed under non-ideal conditions might be subject to diverging intuitions.

As for deep disagreement and other technical or semi-technical notions, the situation is different. Since these are not common-sense concepts, there are no shared intuitions about which phenomena typically fall under such categories. Consequently, there are also no shared intuitions about paradigmatic examples of deep disagreement. This is manifest in the debates over deep disagreement. Fogelin (1985) presents moral philosophical disputes about abortion and affirmative actions as examples of deep disagreement, and Feldman (2005) thinks that deep disagreement typically concerns deep philosophical questions. In contrast, many authors nowadays regard forms of ‘ideological’ disagreement, e.g., between a creationist and a scientist about the age of the Earth or between a conspiracy theorist and a non-conspiracy theorist, as paradigmatic. However, there is no criterion for deciding the question about paradigmatic examples of deep disagreement, since there is no common concept or notion about which shared intuitions exist. Hence, different theories about deep disagreement will lead to different views about paradigmatic examples and vice versa without any agreed upon mechanism for testing the theories against our intuitions.

There is a further peculiarity of analyzing deep disagreement in the same way that we do concepts such as knowledge, belief, or truth. In these cases, we start from a common concept that we analyze, and typically we assume that there is a single phenomenon, or a specific type of phenomenon, associated with it. For example, Williamson (2000) assumes that knowledge is a specific type of factive mental state, and Sosa (2007) argues that knowledge is a certain type of capacity to competently acquire true beliefs. Thus, when analyzing stable common-sense concepts, there is a reason or a tendency (though not a strict rule) to assume that these concepts refer to a single phenomenon or a type of phenomenon.

The situation is different with deep disagreement because we do not have any common-sense concept or notion as a starting point. Thus, there is no reason to assume that there is a single phenomenon, or a specific type of phenomenon, to which philosophers, linguists, and argumentation theorists refer with the phrase “deep disagreement.” Accordingly, there is no reason to assume that there might not be a variety of forms of disagreements that are “deep” for very different reasons that are then rationally resolvable (or rationally irresolvable) for very different reasons.

Notably, such a pluralistic approach is, to a certain extent, rejected by the metaphysical program of deep disagreement. Advocates of hinge theories and advocates of first epistemic principle theories both tend to provide a theory about the nature of deep disagreement, proposing their own standpoint as correct and alternative ones as incorrect. However, in proceeding this way they tacitly assume that deep disagreement is a single (natural) phenomenon. This strategy is plausibly based on analyzing deep disagreement analogously to common sense concepts such as knowledge, but this strategy is misleading because deep disagreement is not a common-sense concept.

6. A similar problem for an alternative methodological approach

I have argued for the view that the current debate about deep disagreement is methodologically flawed because it aims at analyzing the concept of deep disagreement as if it were a common-sense notion even though it is a technical or semi-technical term. One might suggest a more charitable interpretation of the current debate based on Fogelin's original work in which the debate is less metaphysically loaded and therefore does not suffer from the methodological flaws of the metaphysical project.¹⁶ Following this interpretation, alternative theories do not aim at analyzing the concept of deep disagreement but at developing Fogelin's analysis of a particular argumentative phenomenon. Accordingly, the current debate takes two of Fogelin's stipulations about deep disagreement as a starting point, namely that deep disagreements occur in situations where two parties attempt to use argumentation for persuasion but the parties lack a necessary shared background of beliefs and preferences. Using this understanding of deep disagreement, hinge theories, and first epistemic principle accounts can then be interpreted as alternative views about the nature of the missing shared background beliefs whose absence explains the target phenomenon. In this case, some core features of the phenomenon identified by Fogelin are assumed to explain other features, like the intractability of deep disagreement. We can then consider whether deep disagreement is rationally resolvable, what the potential domains of deep disagreement are, and whether the target phenomenon is something like a natural kind. According to this picture, it is an empirical question whether the instances of disagreement considered actually involve hinge propositions or first epistemic principles.

The current literature does not explicitly refer to this methodological approach, but perhaps this is (at least in some cases) the underlying motivation. However, for the sake of argument, it is worth considering this alternative approach. At first glance, this approach does not face the methodological defect of analyzing the technical or semi-technical term of deep disagreement as if it were a kind of common-sense notion. Nevertheless, I have reservations about this project. The phenomenon to be explained is that two parties attempt to use argumentation for persuasion but lack a shared background of beliefs and preferences. But if this is right, I do not see a reason why there cannot be disagreement based on hinge propositions and disagreement based on first epistemic principles. Accordingly, hinge theories about deep disagreement and first epistemic principle theories just offer accounts of different types of disagreement, which are both instances of the target phenomenon to be explained. However, these theories are often framed as offering competing analyses of the target phenomenon, and if that is the case, then I do not see a criterion to decide which account offers the better analysis except by appealing to pretheoretical intuitions about the metaphysical nature of deep disagreement. In this case, however, one gets the same methodological flaws of misled conceptual analysis. Hence, by considering this alternative approach, we can diagnose a certain persistency of the methodological problem of analyzing deep disagreement.

Moreover, the suggested approach somehow randomly builds on Fogelin's work. Fogelin's paper contains various incompatible characterizations of deep disagreement.¹⁷ First, he introduces the phenomenon of disagreements involving argumentation but lacking a shared background. Second, he presents two philosophical disputes as examples of deep disagreement, disputes about abortion and affirmative action. Third, he claims that deep disagreement is rationally irresolvable. These claims stand in tension. First, Fogelin's examples of deep disagreement are dubious given his own view about deep disagreement, as it is implausible to claim that there are no shared assumptions in philosophical disputes. For example, conflicting parties about abortion or affirmative action can plausibly appeal to more fundamental common moral principles. Accordingly, the current discussion does not regard the cases that Fogelin presents as proper examples of the phenomenon that he introduced. Second, Fogelin merely claims that deep disagreement is rationally irresolvable but does not argue for this claim. However, it is an open question whether disagreements involving argumentation but lacking common background beliefs and preferences are rationally irresolvable as Fogelin claims. Accordingly, it is debated whether those philosophical disputes, which Fogelin regards as instances of deep disagreement, are really rationally irresolvable.

Given these tensions, there are various possible starting points for fixing Fogelin's incoherent project. First one might start, as suggested, by acknowledging the phenomenon of argumentation that lacks shared beliefs and preferences and then inquire about its rational resolvability among other issues. Second, one might use Fogelin's examples of philosophical disputes and then raise the question of whether they are deep and/or rationally irresolvable. This is the path taken by Feldman (2005). Third, one might take the idea of rationally irresolvable disagreement as a starting point for developing a theory about this phenomenon, as I propose in Melchior (2023b). Given the tensions in Fogelin's own work, I do not see a reason why the first project should be more in the spirit of Fogelin's work than the other two.

7. How to do better

Deep disagreement is not a common-sense concept like knowledge, belief, or truth. Accordingly, we cannot analyze it in the same way we analyze these other epistemic concepts by comparing them with our pre-theoretical intuitions. In this respect, the current metaphysical program of investigating the nature of deep disagreement by providing a conceptual analysis, a descriptive definition, or a Carnapian explanation is methodologically mistaken.

At this point, we have two options concerning the notion of deep disagreement. We can, despite the methodological shortcomings, stick to the usage of the notion, or we can abandon it. Which option to choose depends on the practical advantages of having a notion of deep disagreement. For example, using such a notion can be convenient when it brings together numerous phenomena under a single umbrella term. On the other hand, using such a notion is misleading if it is not sufficiently clear what the commonalities of the subsumed phenomena should be. With deep disagreement, there is no consensus about the nature of deep disagreement, what the paradigmatic cases of deep disagreement are, or whether it is rationally resolvable. Accordingly, there are multiple disagreements, and potential confusions, about deep disagreement itself. Thus, the disadvantages of holding on to the notion of deep disagreement outweigh the potential benefits of uniting various different phenomena under one umbrella term. Consequently, I propose eliminating the notion of deep disagreement from the philosophical discussion.

Arguing for abandoning the notion of deep disagreement is a form of eliminativism about deep disagreement. However, this form of eliminativism differs crucially from other versions that have been proposed in the philosophy of mind and thus does not suffer from the well-known problems of these views. Traditional eliminative materialism in the philosophy of mind, as defended by Churchland (1981), Churchland (1986), or Stich (1983), has it that our ordinary concepts of mental states such as belief or the mind do not refer to any particular entity. Consequently, ordinary law-like statements about mental states, like that certain beliefs and desires jointly cause certain actions, should be abandoned. Eliminative materialism is a radical and controversial view since it suggests doing away with deeply rooted, and perhaps even innate, common-sense concepts about mental states. The notion of deep disagreement, which I am currently suggesting should be eliminated, is a philosophical notion or a technical or semi-technical term that gained some popularity in philosophy, but which is by no means an ordinary common-sense concept. For this reason, eliminating it is far more uncontroversial and does not conflict with our common-sense intuitions.

How, then, should we proceed in analyzing those interesting forms of disagreement that are often subsumed under the label of “deep disagreement”? We should start by acknowledging their heterogeneity and consequently by carefully distinguishing them.¹⁸ Accordingly, advocates of hinge epistemology should formulate forms of disagreement based on their particular understanding of hinges, and adherents of first epistemic principle theories should spell out what first epistemic principles are and then characterize forms of disagreement based on these principles.

Are these forms of disagreement rationally resolvable? This depends on the propositions about which the parties disagree and on the concept of rational resolvability in place. Thus, a further central task is to develop a theory of rationally irresolvable disagreement. Here, different forms of rational resolvability and irresolvability will have to be to distinguished since rational irresolvability is a multiply ambiguous notion. As Ranalli notes (2021: 985): “We might have in mind that (i) there's no epistemically rational way to resolve the disagreement; or that (ii) there's no practically rational way to resolve the disagreement; or further still that (iii) the disagreement isn't resolvable by way of argumentation and the exchange of reasons (e.g., paradigmatic rational methods or ways of resolving disagreements.)” I develop such a full-fledged theory of rationally irresolvable disagreement in Melchior (2023b). I distinguish between rational irresolvability in an objective sense and in a subjective sense. Disagreement is rationally irresolvable in an objective sense if there is no objective rule of rationality available such that two disagreeing parties can resolve their disagreement by following this rule. It is rationally irresolvable in a subjective sense if two parties who are following their own beliefs about rationality and evidence cannot reach an agreement because they fail to share sufficiently overlapping views about rationality and evidence. While objectively rationally irresolvable disagreement is irresolvable for objective reasons, its subjective counterpart is irresolvable for subjective reasons. However, these subjective reasons are not purely psychological reasons, such as stubbornness. They are systematic in the sense that the disagreeing parties systematically follow their beliefs about rationality and evidence, even when these beliefs might be false. I argue that, in the context of discussions about deep disagreement, only rational irresolvability in a subjective sense is crucial. In contrast, authors such as Feldman (2005) and Matheson (2021) who argue for the rational resolvability of deep disagreement apply a notion of objectively rational resolvability.

I also discuss in Melchior (2023b) the relationship between theories of deep disagreement and of rationally irresolvable disagreement, ultimately arguing for replacing discussions about deep disagreement by theorizing about rationally irresolvable disagreement. Admittedly, the presented arguments are not conclusive, and combining theories about deep disagreement and rationally irresolvable disagreement is a viable option. Given the inherent methodological flaws of the metaphysical program that we have discussed, however, combining these theories is not an attractive option, since theorizing about the metaphysical nature of deep disagreement is itself flawed.

The alternative methodological account proposed here allows us to distinguish different forms of disagreement and of rational resolvability. One might object that this can also be achieved on the basis of the criticized metaphysical program. Moreover, this program is compatible with various forms of rational irresolvability. Thus, one might not see any methodological flaws in continuing the metaphysical program. However, pursuing this program also means to further investigate which of these disagreements are deep and which ones are not. In contrast, the alternative approach proposed here recommends that instead of discussing the real nature of deep disagreement or searching for the depth of deep disagreement, we should only identify and distinguish various forms of disagreement that are fundamental in various ways. The metaphysical program of deep disagreement is methodologically flawed in contrast to the account proposed here.

8. Conclusion

One enterprise concerning deep disagreement is the metaphysical program of inquiring into its metaphysical nature. This project is methodologically flawed, since deep disagreement is neither a natural language notion nor a common-sense concept but rather a technical term. Consequently, there are no pre-theoretical intuitions about deep disagreement available for testing theories about deep disagreement. In this respect, anything goes for theories of deep disagreement. Moreover, deep disagreement is not a useful umbrella term that subsumes various related phenomena. Rather, it creates confusion and misunderstanding along a number of dimensions. Therefore, we should eliminate the notion of deep disagreement from philosophical discourse and instead theorize about different forms of disagreement, namely disagreements involving hinges, disagreements involving epistemic principles, and rationally irresolvable disagreements.¹⁹