Deep Agreements

Abstract

Recent research has highlighted the character and importance of the study of agreement. This paper, paralleling work on the more familiar concept of deep disagreement, will provide a first articulation of the character and implications of deep agreements, that is, agreements so deep that disagreement cannot overcome them. To do so, I start by outlining the main features of deep disagreement. I then provide a brief characterization of agreement in general to ground the discussion of the unique characteristics of deep agreements and compare them to deep disagreements. After distinguishing the concept from other more familiar ones, I then point to a few of the major social and theoretical issues the existence of deep agreements poses before concluding with some remarks pointing to valuable areas for future research.

1. Introduction

When Jordan Klepper, a comedian on “The Daily Show” who is known for conducting brief field interviews with Trump MAGA¹ supporters, was asked about one of his most memorable moments, he singled out one specific interview he conducted during the first Trump impeachment. He explains,

I was talking to a woman at the first Trump impeachment, and Trump was blocking witnesses from testifying…he was blocking [former national security advisor] John Bolton from testifying – among others. And, I was talking to this woman and she said, like ‘He's innocent! He's completely innocent! He didn't do anything wrong. If he was trying to do…if he did something wrong, he would be trying to hide it.’ And I was like, ‘Well if he was stopping people from testifying, that would be an admission guilt, right?’ And she said ‘Of course! Of course!’ And then I told her, I was like, ‘He is blocking people from testifying!’ And she takes this very long beat. She thinks about it. And she says, ‘I don't care.’ She was being completely honest in that moment. (Klepper 2023)

Klepper's experience here, and the unidentified woman's reasoning, are both fascinating and could be explored from several perspectives for several purposes. The most basic level of analysis must point out the expectation that the woman, faced with such a self-contradiction, would retract her position that Trump is innocent. Given the claims Klepper presents her with, and her reluctance to retract her position, one might be tempted to simply evaluate this situation as an instance of manifest irrationality. Leaving the evaluation there would, however, miss a central component of what is happening, and one that may explain (justification being a separate issue) why the woman responds the way she does. In this paper, I contend that this example may point to what I call a ‘deep agreement’, that is, an agreement so deep that disagreement cannot prevail. More specifically, I will argue that it may serve as an example of what I call an entrenched agreement, and I will then connect that notion to Klepper's later observation about this woman's identity as MAGA.

To do so, I start with a brief discussion of a more familiar concept: deep disagreement (section 2). I then provide an overview of my previous characterization of the nature of agreement in general (section 3) to help the extrapolation of two varieties of deep agreement (section 4). To help illustrate the uniqueness of the phenomena, section 5 distinguishes deep agreement from related concepts. The penultimate section (6) will review some of the impacts that deep agreements have on both the social and theoretical levels, while the conclusion (section 7) will summarize the arguments presented and point towards avenues for future research.

2. Deep disagreement

Although this paper is focused on deep agreements, many readers have likely thought much more about the well-known phenomenon of deep disagreement, so it is worth briefly reviewing this more familiar concept at the outset to help ground a comparison of deep agreement and disagreement below.²

Research on deep disagreement has greatly expanded since Fogelin's ‘The Logic of Deep Disagreement’ first appeared in 1985³ (e.g. Davson-Galle 1992; Kappel 2018; Lavorerio 2020, 2021; Lugg 1986; Ranalli and Lagewaard 2022a, 2022b; Turner and Campolo 2005). One main, perhaps unintended, result of this scholarship has been to demonstrate that despite widespread intuitions about the existence of the phenomenon, finding a commonly accepted definition of it has been incredibly difficult. To ground this discussion, I will start with a short review of Fogelin's central points before briefly reviewing some commonalities in the research that have emerged since then.

For Fogelin, who credits his insights to Wittgenstein's On Certainty (1969), deep disagreements ‘undercut the conditions essential to arguing’ (2005: 8), which makes them impossible to resolve. They undercut these conditions because they stem from an abnormal argumentative context wherein two features stand out. First, there is a lack of broadly shared beliefs and preferences between the interlocutors. Second, but relatedly, there are no shared procedures for resolving disagreement. In other words, there is a significant mismatch in both the form and content of the argumentative environment. In a ‘normal’ argumentative context, participants have both broadly shared beliefs and preferences, and procedures for resolving disagreement. Thus, for Fogelin (2005: 7), ‘to the extent that the argumentative context becomes less normal, argument, to that extent, becomes impossible’.

In the years that have followed Fogelin's initial articulation, scholars have advanced the literature on deep disagreement by working to refine the characterization/definition of the phenomenon and providing methods for its possible resolution (if optimistic) or arguing for its intractability (if pessimistic about the possibility of resolution; see Mota forthcoming [2025]). As Lavorerio (2021: 418) astutely points out, deep disagreements are often persistent, lack a clear path towards resolution, have a ripple effect, are emotionally charged, and involve a difference in worldviews. Duran (2016: 4), while discussing possible levels of depth in deep disagreements, notes that Davson-Galle (1992), Turner and Wright (2005), Memedi (2007), Campolo (2005) and Friemann (2005) all employ the term ‘intractable’ (be it potentially, seemingly, or completely) while discussing Fogelin's views and suggests that it may be a better term than ‘deep’. Since the notion of resolvability/intractability has continued to occupy a central place in the discussion (e.g. Ranalli 2021), it stands out as one generally common thread within the scholarship and will be important for the discussion of deep agreements below.⁴

For several scholars, what makes these disagreements persistent and (definitely/potentially/seemingly) intractable is that, as Fogelin also pointed out, they involve content deeply embedded and interconnected into the web of beliefs that contribute to the make up of our identity. While there are theoretical differences between the various characterizations and names given to this content, for our purposes it is enough to note that they are most often referred to as framework propositions (Fogelin 2005; Lavorerio 2020), hinge propositions (Coliva and Palmira 2020, 2021; Johnson 2022; Pritchard 2018; Siegel 2021), worldviews (Lavorerio 2021) or structuring commitments (Mota forthcoming [2025]). To illustrate the embedded nature of these kinds of propositions and how they often go unscrutinized, Wittgenstein provides the following example:

I have a telephone conversation with New York. My friend tells me that his young trees have buds of such and such a kind. I am now convinced that his tree is… Am I also convinced that the earth exists? The existence of the earth is rather part of the whole picture which forms the starting-point of belief for me. Does my telephone call to New York strengthen my conviction that the earth exists? Much seems to be fixed, and it is removed from the traffic. It is also so to speak shunted onto an unused siding. (1969: 208–10)

The example shows how the conversation about buds on a tree presupposes the existence of the earth but does not seem to count as direct support for that broader claim. Rather, that the earth exists, as a commitment, is left unquestioned (‘shunted onto an unused siding’). To see the embeddedness of the proposition that the earth exists, one need to only think about how many of their other beliefs would have to change if such a proposition were thought to be false.

In sum, then, deep disagreements are commonly thought to be intractable because they involve the contrast of differing structuring commitments, which are embedded and widely entangled into the web of beliefs that contribute to the make up of our identity. In some cases, they may also involve a lack of agreement on acceptable procedures for resolving the disagreement.⁵ Nevertheless, combined, these conditions may undercut the conditions necessary for argumentation to effectively resolve a difference of opinion.

3. Agreement in general

So, what then of agreements? Before characterizing deep agreements and comparing them to their disagreement counterparts, it is important to characterize agreement generally so that the separation between ‘normal’ and ‘deep’ agreements becomes more visible. In a first paper on this topic, I recently characterized agreement as ‘the recognition of an implicit or explicit, mono, bi-, or multi-directional congruency of an attitude or attitudes held by interlocutors or within ideological positions’ (2023: 269). More simply, as ‘the recognition of congruency of an attitude or attitudes, which in argumentation appears to come in five varieties – agreement that/to, agreement because, agreement so, and agreement with, – all of which fall along a scale of degree’ (Baumtrog 2023: 276).

The two central features of this characterization are kind and degree. In terms of kind, agreements ‘that’ or ‘to’ point to agreements on a conclusion – either of a belief or a course of action (intention to act), respectively. An agreement ‘because’ is an agreement on the reasons for the conclusion. Agreements ‘so’ occur when agreements that or to are arrived at to facilitate a further end, such as in a mediation. The most complicated type of agreement pointed to in this article is the agreement ‘with’. It is most complicated in part because it points to two distinct notions of agreement with. In the first sense, the ‘with’ simply identifies a target of agreement, most often a person (e.g. I agree with my boss that we should hire Shareena). In the second sense, however, the agreement with identifies a broad or loose content, such as an ideology (e.g. I agree with free market capitalism; I agree with socialist democracy).⁶ This second sense is left vague because there may not be any way to specify an exact list of propositions equal to the ideology or position being identified with – nor may such a list be needed for the agreement to make sense. In this way, one can sensibly say they agree with socialist democracy, while maintaining some more specific disagreements with others who similarly identify.

In terms of degree, however, in that earlier work I avoided a developed discussion of the depth of agreement but did point out that agreements come in degrees. Using the example of planning a trip to Italy with my wife, I explained that ‘I may agree with my wife generally and agree with her idea to visit Italy, but whereas 100% of her reason to visit is because it allows her to see her parents and she has no interest in visiting the beach, I may be evenly split between the two. Thus, when searching for agreement on the meta level, we can find agreement in both type and degree’ (Baumtrog 2023: 272). Given the definition of agreement I provided there, the term ‘degree’ seems to be referring to the degree of congruence or overlap in the types of agreement – from zero to total overlap. But there is another, perhaps more common sense of ‘degree’ that relates to the strength of the conviction in the agreement, which I did not specify previously. Developing this notion leads us to the idea of a deep agreement, which I turn to next.

4. Two varieties of deep agreement

Returning to the opening example of the MAGA supporter, what can we say about the type of agreement that she seems to maintain despite the demonstration of evidence that was expected to have produced a change of mind? In other words, why might the woman have responded with ‘I don't care’, signalling her reluctance to change her mind and her preference to continue her agreement with MAGA? Using the framework from above, the first thing to note is that the woman does not maintain all her instances of agreement with MAGA. In fact, her response of ‘I don't care’ seems to signal an acknowledgement (if not agreement) that stopping people from testifying fits her previously agreed criteria for ‘hiding something’, and that Trump in this case was stopping John Bolton from testifying, thereby constituting a cover-up. That she provides no other reasons to oppose the notion that a cover up was happening leaves her at least in the space between agreement and disagreement (but she seemingly concedes) that something was being hidden. However, ‘I don't care’ also purposefully indicates the maintenance of some component of her previous position.

In terms of the types of agreement covered above, despite her seemingly conceding that Trump was blocking people from testifying, I contend that she is still maintaining/refusing to give up her broad agreement with the ‘MAGA movement’. Recall that one aspect of agreement with an ideology is the breadth of the agreement it manifests, such that one can agree with the ideology or position broadly despite potentially imperfect agreement on some of the specific or possible propositions that make it up (if such a list could be formulated). Thus, what we see here is a display of the woman's continuing deep agreement with the MAGA ideology/movement through her reluctance to give up her claim regarding Trump's innocence more broadly even if she gives up the more specific that claim regarding Trump blocking testimony.

Together, these observations demonstrate two important features of deep agreement. First, that the woman could drop one point of possible agreement while maintaining another shows that deep agreement does not have to cover more than any one type of agreement – its most important feature is in degree of conviction. For a deep agreement to exist, one need to only unshakably/intractably agree that, to, because, so, or with. However, when someone deeply agrees with more than one type of agreement or more than one instance of the same type of agreement, we can specify that the agreement is becoming entrenched, which is to say, both wide in types or instances and deep in degree. Insofar as agreement with the MAGA identity can be broken down into agreements that, to, etc., we can see how removing one root of agreement – e.g. that Trump was admitting guilt – is not enough to move the rest of the entrenched agreement composing the broad agreement with the MAGA identity, and instead allows the interviewee to maintain her alliance to her identity and community.

Klepper also recognizes this. Shortly following the quote provided at the beginning of this paper, he continues:

And she's right. So much of this we're like ‘Oh but maybe I could convince that person. That they could care if there was that one thing.’ I read something recently that said like, you know ‘We can have debates about what you want. That's politics – I want this. I want that. Let's meet somewhere in the middle.’ That's politics. When your politics becomes who you are, we can't debate that. Real hard to change. And the Trump magic trick is that he made folks think that this is who they are. They are MAGA. So, when we talk about ‘Do you think he's guilty? Do you think he's not guilty?’, we can banter back and forth about the details, and we can quibble in that way, but we're not really coming to an understanding of truth. That's who that person is, and they don't give a shit about the new piece of information. And so that moment for me was revelatory. (Klepper 2023)

The agreement here is with an ideology, with an indefinite constellation of propositions (and likely non-propositional content), and is held with deep conviction – so entrenched that it has become a central component of the woman's identity. We do not need to see a list of all the propositions that make up the MAGA ideology to see this, nor do we need to test each one for depth. The woman could also readily disagree with others who identify as MAGA on any given proposition included in the set of beliefs that help constitute the ideology and still understandably express agreement with it.

Second, but relatedly, this example shows that having many points of agreement is not necessarily enough to constitute a deep agreement. This is because you could have complete overlap on multiple agreements that, because, etc., but also be readily persuadable for each one of them. One common example might include working with a colleague to plan an event. You and your colleague might agree to hold the event on date x, because of reasons 1, 2, 3. And you might have 100% overlap on all those things, but also both be easily persuaded to change both the date and reasons if presented with a (some) compelling counter-consideration(s). Thus, if we think of deep agreement like an immovable fence post, and entrenched agreement like the many widespread and deep roots of an old tree, we might think of a merely wide agreement as a patch of grass. And like a patch of grass, even though it may cover a broad area, it is often easier to uproot, roll, remove, and replace than a tree or grounded fencepost. Thus, not all agreements with, even when they are with broad ideologies, are deep.

In light of the overview of deep disagreement presented above, we can now see how the general notion of deep disagreement relates most clearly to my articulation of entrenched agreements. It seems that if two people with entrenched agreements came into conflict, it would necessarily entail that they are in a deep disagreement – as they have been commonly characterized – at least in content if not also in terms of procedures towards possible resolution.

But what about two people with merely deep agreements? In that case, I do not think a deep disagreement, as traditionally characterized, necessarily results. For example, my roommate and I might hold an unshakable agreement that on each of our respective birthdays we attend a sporting event together, and we do so because it is tradition. In this case we have an agreement to (attend the event) and because (it is tradition), but it is the latter agreement we feel most deeply and that perpetuates the former. However, the agreement to follow this birthday tradition does not seem to be part of any entrenched agreement in that it is not a by-product of a more entrenched agreement on the value of tradition generally, or that manifests in other parts of our lives. In fact, aside from this, we each keep very few traditions. But when faced with alternative plans presented by our other friends who all deeply agree that we should do something else on our birthdays, we never accept attending alternate events because of our deep agreement regarding the value of this tradition. In this way, two merely deep agreements do not seem to include worldviews or structuring commitments as the deep/entrenched disagreement literature has traditionally discussed.⁷

Moreover, mere doubt or opposition to an entrenched agreement may also be insufficient for causing a deep disagreement. For example, I may have an entrenched agreement with an ideology, MAGA or otherwise, and someone could understandably hold that I should not agree with that ideology as deeply as I do, but that does not mean we have a deep disagreement. It may mean that we have ineffective communication, or that one of us is being overly stubborn. It may also simply mean we dislike each other. But the presence of mere doubt does not seem to undercut the possibility for resolving the disagreement (create a deep disagreement) in nearly the same way as deeply holding an opposed structural ‘agreement with’ – an entrenched agreement – and having it come into contact with its opposition would constitute a deep disagreement.

We are now able to identify some possible components in the start to a typology of degrees of agreement (summarized in Table 1 below). First, as was illustrated using the example of conflicting birthday plans, there are ‘merely deep’ agreements, the main feature of which is the depth of shared conviction regarding the agreement. As a simile for this singular depth, I have suggested thinking about them as a sort of deeply buried, singular, fencepost. Second, and much more closely paralleling the literature on deep disagreements, are what I am calling entrenched agreements. Much like the roots of an old tree, entrenched agreements involve several deeply held agreements, some of which are more central and important to the maintenance of personal identity (the tree) than others. Thus, the roots can be likened to the intricacies involved in systems of worldviews, hinge propositions, structural commitments, and the like. Third, but only as a matter of contrast, I also introduced the notion of wide agreements using the simile of a patch of grass. Like the many shallow roots holding down a patch of grass, wide agreements exist when there are several points of agreement (of the same or differing types), but all or most could be easily overcome.

Table 1. Three degrees of agreement.

To conclude this section, it is also worth suggesting how these metaphors may help point to possible methods for breaking up deep agreements if they should be so broken up (as discussed below). For example, while there may be some similarities in how you remove a fencepost and an old tree, there will also be important differences. Whereas digging deeply straight down might be effective for removing a fencepost, to remove an old tree it may be best to first starve the roots of water. Similarly, while it may be best to address a merely deep agreement with a dedicated and deep diving discussion on the topic, to remove an entrenched agreement it may be best to change the discussant's environment (remove the usual water source) before digging (or discussing) either around the roots or closer to the tree trunk. But these are just speculations. While my focus in this article is only to make the case that deep and entrenched agreements exist and deserve philosophical attention, future work could explore the possible methods for disrupting or dissolving deep or entrenched agreements along these lines.

5. Relationship to related concepts

By this point, one might have had the thought that my notions of deep and entrenched agreement are just re-wordings of the more familiar concepts of belief, closed-mindedness, or indoctrination. While a full juxtaposition between the concepts is beyond the scope of this paper, it is worth quickly clarifying why I think these concepts are substantively distinct.

5.1. Belief

Beliefs are necessarily related to agreements ‘that’. On my earlier account, to agree that X is for two or more people to share (maintain congruency with) a belief, or for one person to share a belief that is expressed through an ideological position. It is similar for intentions, i.e. agreements ‘to’. On one understanding of intention, to agree to X, is for two people to jointly form the intention to X. However, discussion of belief, especially in relation to conviction, is most often discussed on an individual level.⁸ One thing this discussion of agreement has brought to the fore is the complication of adding another person or perspective into the calculation of the degree of overlap or the overlap in conviction of belief (in the cases of merely deep agreements on single propositions) and the expansion of agreement to an indiscernible set of propositions that help constitute an ideology or identity (in the case of wide or entrenched agreements). In other words the focus here has shifted from ‘I believe that X’ to ‘We/I agree that x, y, z….n…(?)’, where ‘x, y, z…n’ are a collection of discernible propositions and ‘(?)’ is the indiscernible constellation of considerations that help constitute an ideology or identity. Thus, for analytic purposes, there will be some cases where discussion of agreements can be helped by looking for individual beliefs, but the story of agreements (deep, entrenched, wide, or otherwise) will be left incomplete without considering the impact of other people, propositions, and convictions.

A non-political example of entrenched agreement may also help clarify here. Consider, for example, the difference between scientific advancements and paradigm shifts. Both advancements and paradigm shifts involve a change of belief, but the latter are only possible when the change in belief is both (a) widespread and (b) a change so central that it has a wide-ranging impact on other beliefs. That it is widespread means that many others agree with it, and that it impacts so many other beliefs, shows its embeddedness. A scientific advancement does not need nearly the same extent of agreement or impact on other beliefs.⁹

5.2. Closed-mindedness

Unlike beliefs, closed-mindedness is both distinct from and unnecessary for deep agreement. First, in agreement with Battaly (2018b: 24; see also Battaly 2018a), ‘closed-mindedness doesn't require having extant beliefs about the given topic’. If it does not require extant beliefs, I hold it also does not require extant agreements. In other words, closed-mindedness can exist before agreement ever comes into the picture. One may be closed-minded to agreeing that, to, with, etc., and thus never even start a process of coming to an agreement.

Second, closed-mindedness is unnecessary because even the most open-minded people could come to deep agreements. In other words, closed-mindedness is unnecessary because deep agreements can exist without it. In fact, open-mindedness may actually strengthen a deep agreement. If, for example, two people conduct an ideal instance of argumentation, say a pragma-dialectical critical discussion (van Eemeren and Grootendorst 2004 – discussed below), and through that procedure of critically testing a standpoint settle on an agreement that X, the inability to find persuasive counter-considerations, especially after doing everything possible to find them, would typically strengthen the agreement rather than point to a closed-minded acceptance of the conclusion.

5.3. Indoctrination

It is unsurprising, but telling for our purposes, that the main locale for philosophical discussions of indoctrination is within the philosophy of education. It is telling because like education, indoctrination tends to identify a process rather than a result. This is the first way that deep or entrenched agreement differs from indoctrination. While a deep or entrenched agreement is something we can point to (even if it evolves, deepens, or shortens), indoctrination is not a result in the same way. So, while indoctrination may lead to a deep or entrenched agreement, it is not the same thing as a deep or entrenched agreement.

Second, and more importantly, it seems entirely possible to indoctrinate someone into an ideology that you do not hold, but impossible to be in a deep agreement with someone on a belief you do not hold. I am thinking here of stories like in the Bourne franchise, where Jason Bourne is originally indoctrinated by government operatives to become an assassin. In a case like this, those who performed the indoctrination may not hold the same beliefs he is indoctrinated into but perform the indoctrination anyway so as to achieve alternate ends. Finally, as with closed-mindedness, the opposite of indoctrination (say, critical thinking), may serve to strengthen/deepen an agreement.

If I have been accurate thus far in identifying deep and entrenched agreements as unique phenomena, perhaps connected to but not synonymous with these related concepts, then it is worth thinking a little bit about what the implications of this phenomenon may be for our lived experience as well as for argumentation theory, where agreement and disagreement take centre stage. I turn to these topics next.

6. Some social and theoretical implications

6.1. Social implications

Given the nature of the MAGA example used throughout the paper thus far, the most obvious social repercussion of a deep or entrenched agreement occurs in the political domain. If democracy ought to be deliberative (Bohman and Rehg 1997; Dryzek 2010; Gutmann and Thompson 2004), meaning broadly/roughly that political/societal decisions ought to be based on the ‘unforced force of the better argument’ (Habermas 1996: 306), competing groups of people who are certain that their decisions are the result of the better argument may become unmovable. In other words, deep agreements can significantly hinder the ideal of deliberative democracy. This does not necessarily mean that democratic deliberators are closed-minded (though they surely could be!). As mentioned above, a deliberative democrat could in good faith consider all available reasons, and feeling strengthened by their due diligence of seeking out all available reasons develop a deep agreement with other like-minded individuals about the best course of action. The problem, then, is not with their closed-mindedness towards available evidence or reasons but how they evaluate them both in isolation and as incorporated into their overall political ideology. But in the absence of a standard formula for determining the better argument, deliberators can understandably differ in their assessments. In this way, as we saw with the opening example, even the removal of one (deep) agreement may not be enough to move an entrenched agreement under consideration in a deliberative forum.

Relatedly, recall that the main example from this paper shows how the agreement with applies in two ways, in that Trump (as a target of agreement – ‘agree with Trump’) is also the main hub or spokesperson for the general MAGA framework that is also being agreed with (agree with MAGA). As a spokesperson, Trump sends the heart of the MAGA ideology out into the world. However, as a representative in a representative democracy, Trump is also a reflection of the ideology of the community. In this way, the my-side bias (Mercier 2022) runs rampant in that Trump reflects the people and the people deeply identify (maintain entrenched agreement) with Trump. This reciprocal relationship may have a notable impact on entrenching the agreement and undercutting the conditions required for deliberative democracy. In this way, as John Casey notes, ‘agreement, consensus, and collaboration can function as means of control, manipulation, or domination’ (2020: 104), which I think is more likely when the agreement is deep or entrenched without ethical regulation.

While these worries may seem empirically rare, increased attention to extremist violence may suggest that they manifest more commonly than we suspect. Consider, for instance, the role that the creation and attempted dissemination of a manifesto played for Ted Kaczynski, the Unabomber, and Anders Breivik, the 2011 Norwegian mass murderer. In both cases, these violent extremists felt it important and necessary to take their views, themselves a likely product of previous entrenched agreement, and disseminate them in such a way that they would gain the entrenched agreement of as many others as possible. For Kaczynski, the attempt to gain widespread agreement was made through a combination of his violence and the publication of his manifesto in a major national newspaper – the Washington Post. Approximately 15 years later, after the internet had become a fixture across large parts of the world, Breivik used it to explicitly address those with whom he already felt he maintained entrenched agreement and to gain the widest number of entrenched agreements possible among people unfamiliar with his hinge propositions. During a psychological interview (Husby and Sørheim 2011: Sec. 5.4) Breivik stated that his violence at ‘Utøya island and the government building was all about publishing the manifest, to reach the 350,000 militant nationalists who are the audience’. And early in that manifesto he writes, ‘The importance of spreading the truth and distribute (sic) sound strategies cannot be underestimated as it is at the very core of our current resistance efforts’ (Berwick 2011: 8). It was at the ‘very core’ because he was sure that ‘If you read it from the first word to the end, you will be radicalized’ (Husby and Sørheim 2011: Sec. 5.4).

Two final notes are crucial to add to these first thoughts regarding some of the societal implications of deep agreement. First, while in this paper I have highlighted an example using Trump and the MAGA ideology, along with two violent extremists, all of whom are most often associated with right or far-right wing politics, it is worth making explicit that deep agreement is not a partisan phenomenon. We could readily see a similar phenomenon happen with Stalinists, most often described as far-left wing. This points to the observation that a notable figurehead for an ideology may be as important to spurring deep agreement as the content of the messages and ideologies they espouse.

Second, and similarly, deep agreement can be, but need not be, harmful or irrational.¹⁰ For example, I struggle to think of a harm resulting from a deep agreement regarding women's suffrage (a deep agreement that women should have the right to vote), or an entrenched agreement about women's equality more broadly (with the ideology that women and men should be considered equal in all possible ways). Thus, I do not want to be taken in this paper as saying that all deep or even entrenched agreements should be avoided. Which ones should and should not be avoided or overcome is an ethical and pedagogical question worthy of investigation. My primary aim here has been to draw attention to the phenomenon and start to characterize it in philosophical terms for future development, critique, and clarification.

6.2. Theoretical implications for argumentation theory

On the theoretical level, the existence of deep agreement raises important questions for some of the most well-known theories of argumentation. For example, according to the pragma-dialectical model of argumentation (van Eemeren and Grootendorst 2004), when performed correctly, argumentation is the ideal method for rationally resolving a difference of opinion. The ideal model proposes four stages and ten rules for a critical discussion. During the fourth stage, participants take stock of the results and assess both the potential for a change in attitude towards the standpoint as well as the belief and degree of commitment towards it. They write:

In drawing the conclusion, it can emerge that the standpoint can be upheld. In that case, the standpoint may be repeated (‘I uphold my standpoint’). It can also happen that the standpoint is to be retracted. Standpoints or arguments can be advanced by assertions but also by other assertives such as statements, claims, assurances, suppositions, and denials. The belief in a proposition and the degree of commitment to the proposition expressed in a standpoint or argument can vary from exceptionally strong, as in the case of a firm assertion, to considerably weaker, as in the case of a supposition. (2004: 63–64)

Since, if disagreement persists after the concluding stage, participants can re-engage in another round of the critical discussion in an attempt to resolve it, it may seem that on the pragma-dialectical model an ideal resolution of a difference of opinion consists of holding as many critical discussions as needed until deep or entrenched agreement is reached. This is because even if the proponent decides to retract their standpoint, they may not do so for the same reasons their discussant thinks they should or may not then hold a counter standpoint to the degree of conviction the opponent thinks they should. In these cases, disagreement persists, and we could envision another critical discussion to address the standpoint that ‘You should strongly believe X, instead of holding it as weakly as you currently do’. Taken to the end, successful critical discussions would end in deep or entrenched agreements.

But van Eemeren and Grootendorst specifically indicate that this should not be identified as the pragma-dialectical aim. They write, ‘It should be borne in mind, however, that the primary aim of a critical discussion is not to maximize agreement but to test contested standpoints as critically as possible by means of a systematic critical discussion of whether or not they are tenable’ (2004: 188). And in a footnote to this passage they clarify, ‘This position could be characterized as “negative utilitarianism.” Rather than achieving the greatest possible happiness, the general aim is achieving the least possible unhappiness’ (188 n. 4).

Still, despite stating that deep agreement is not the aim of the pragma-dialectical model, there is no explanation or rebuttal to the reasoning that might understandably lead people to see it as resulting from the rest of the theory, and that van Eemeren and Grootendorst anticipate themselves. In other words, we are left wondering why differences of opinion regarding the degree of conviction on a standpoint should not be resolved using the pragma-dialectical critical discussion? In another light, it re-raises questions the pragma-dialecticians have faced in some regard before, such as why interlocutors might aim for a satisficing of happiness and tenability, leaving some aspects of doubt or disagreement still in play, especially while discussing matters of truth, such as which city has the highest population density in Europe or the current shape of planet Earth?¹¹ Similar problem can arise with moral conclusions. In other words, why would you not want a goal of argumentation to be for someone to be or become deeply anti-racist, but just anti-racist enough to cross the threshold that they believe racism is not tenable?

Relatedly, on the rhetorical model of argumentation, which holds that one central aim of argumentation is to increase the adherence of an audience to a thesis (Tindale 1999, 2004, 2015), we can now ask how far that understanding and adherence should ideally increase? The question is important because as Tindale writes, ‘Adherence registers the degree of agreement or assent to a thesis’ (2015: 68). Would, on the rhetorical model, deep or entrenched agreement be the ideal goal? If so, what happens to worries about harmful agreement? If not, what is the ideal level of disagreement that should remain?

One potential answer might invoke Perelman and Olbrechts-Tyteca's (1969) notion of the ‘universal audience’, which as an extrapolation from a real audience, is a hypothetical audience imagined by an arguer that can be used to inspire creative thinking but also to gauge the quality of an argument (Baumtrog 2017; Tindale 2004). On this notion, one might hold that the ideal level of agreement is the one that gains the agreement of the universal audience. Appealing to this hypothetical audience, however helpful it may be for improving argumentation, still leaves questions regarding the ideal level of agreement unanswered. For example, how much agreement is envisioned within the universal audience? Should this hypothetical audience be conceived as one that maintains disagreement within its ranks? And if so, to what extent? Then there are further questions about level of agreement between the real interlocutors and the universal audience. Should we aim for entrenched agreement with the universal audience? Would that even be possible? If not, to what extent should they disagree, and at what point would their agreement signal the reasonableness of our conclusions?

Tindale provides some starting points to answers to some of these questions. For instance, regarding agreement within the universal audience he states that ‘it involves no claim to an objective truth that transcends time and place. It involves no necessary sameness of agreement’ (2015: 218). On such a view it seems that disagreement can remain within the universal audience. But nevertheless, Tindale writes, ‘One way or another, “the philosopher must argue in such a manner that his discourse can achieve the adhesion of the universal audience” (Perelman 1979: 58), because the quality of a discourse cannot be judged by its efficacy alone, but also by the quality of the audience on which it is efficacious’ (65).

We see here that both Tindale and Perelman are grappling with a tension between embracing philosophic pluralism (Tindale 2015: 65) while simultaneously centring the importance of the universal audience. And at least Tindale seems to prioritize pluralism at the end of the day: ‘In this sense argumentation might not always be expected to achieve agreement, or even the resolution of disagreements, but the maintenance of diversity in consensual reasoning’ (2015: 196). Thus, as answers to the questions posed above, there seems to be a tension in being directed to aim our argumentation at gaining/increasing the adhesion of (increasing the agreement with) the universal audience, when that audience is one that maintains pluralism within its ranks and when pluralism among real audiences is characterized as so philosophically important. It means we need a much better understanding of good/ideal disagreement, how it fits into argumentation theory, and how to identify when we've taken agreement too far.

Returning to argumentation in practice, as an educator teaching critical thinking to undergraduates, I have found that my answer to these questions has thus far been quite vague and could benefit from clarification: I always want my students to agree with me, but to do so critically. In other words, I consciously want to avoid having them deeply agree, or heaven forbid find themselves in entrenched agreement (however ego boosting that may feel) with me. But at the same time, I want them to sincerely take up and apply the lessons I am working to impart upon them. In this way we can now see the critical acceptance promoted within critical thinking is also implicit within both the pragma-dialectical and rhetorical approaches to argumentation – agreement with room to spare – but the theoretical explanation for this position within argumentation theory still seems to be missing, and at times possible options seem inconsistent with other central components of the respective theories.

7. Conclusion

In this paper, I have offered a first characterization of deep agreements from an argumentative perspective. In doing so, I highlighted the central importance of the depth of conviction involved in the agreement. Given the importance of the level of conviction, an amended version of my initial definition of agreement seems to be in order: the recognition of an implicit or explicit, mono, bi-, or multi-directional congruency of an attitude or attitudes held by interlocutors or with ideological positions, which manifests in degrees of congruency and with varying levels of conviction.

The main benefit of this characterization is that it will allow us to use more precise language to describe and analyse instances of agreement we find in the world. Relatedly, it may also help future characterizations and analyses of instances of deep disagreement.

Beyond the descriptive, however, my hope is that the present characterization will aid future research into prescriptive work addressing the resolution of deep disagreement and the removal of harmful deep agreements. Distinguishing the kinds of deep agreements (deep vs. entrenched) will help point to the differing tools that may be needed in their respective cases just as differing tools are needed to remove fence posts and old trees.

However, determining if a deep agreement needs to be resolved implies an understanding of the ethics of deep agreements. As mentioned, I do not believe all deep agreements need to be broken. If this is correct, it leaves room for discussions of ethical extremism and the role it plays in argumentative interactions.

Finally, my hope is that the discussion of entrenched agreement plays a part in inspiring future research on several more specific topics, like how to dissolve or resolve it. But most importantly, future research could continue to develop the connection between argumentation and identity – a topic gaining most notoriety in Tindale's (2015) work. Since so much of the disagreement literature thus far has focused on single or groups of propositions, more research looking at the connection between the non-propositional components of identity and how it impacts agreement, disagreement, and argumentative discussion would be a valuable contribution.¹²