Reasons For Disagreement

Disagreement is all around us. I don't mean disagreeable people, or those who are not trying to see eye to eye in earnest, but rather, people who truly disagree, or seem to disagree, with one another, or ourselves. Politicians, for example, are forever talking past each other, and at us; but, if we listen carefully to the best of statesmen, we find that they are also disagreeing with each other. Scientists, educators, lawyers, artists, farmers -- no matter who is involved, what medium we reference, what culture we investigate, or in what period a conversation takes place, we encounter disagreement. Seldom, and indeed ever more so, it seems, do we encounter a discussion that is entirely free of disagreement. Even if, for a while, members of a most insular cohort appear to sing in perfect harmony with their songbook's dogma, we know we need merely wait to see the most violent disagreements form within their ranks.

It seems to me that it should be possible to break down and categorize all conceivable sources of disagreement, if only to clarify whether a given disagreement actually has the import it seems to carry, or even if it is worth trying to resolve. High level categorization helps us get a grip on the landscape of interest, and can quickly reveal gaps when our efforts are not exhaustive. Here, I try to break down the reasons for why people disagree, at the highest level that I am able. This, of course, does not mean anything more than it purports to mean; nothing here is definitive, nor is there pretense of this being a singular approach.

The question -- for what reasons do people disagree? -- seems to have a surprisingly short answer. It appears there are only a few categories into which grounds for disagreement fall. Seldom are disagreements so simple as to involve only a single reason or category. But this should not detract from the task of identifying the unique categories that form the grounds for why people don't agree with one another.

1. Informational

Perhaps the most obvious reason for two people to disagree with one another is due to one, or both, being privy to different information. In other words, there is an absence of information for one or both parties. This kind of disagreement should be easily resolvable. Although it is a delight to resolve, it is also quite uncommon.

Another source of informational disagreement concerns the assumptions we make about a given argument. Assumptions can be thought of as unspoken, or invisible information. Examples of this include assumptions about a temporal frame of reference or range; assumptions about the context, or allowable set of contexts, for which the argument is valid; assumption of a moral system in effect (very commonly, utilitarian); assumptions regarding the exact extents of relative words, for instance, "large" or "fast"; or of goals, unspoken and assumed to be common. Since these are merely missing information, they ought just as easily to be rectifiable by challenge and clarification.

A more common cause for disagreement due to informational differences concerns differences in the weights people assign to information. There is no such thing as unweighted information within the human experience. Every "truth" is accompanied by a weight. Children weigh everything they believe they know based almost entirely on experience and trusted authority, and their quickly changing "facts" are an expected result. As one matures, authority is questioned, challenged, and revised -- an especially striking feature of adolescence. New authority, and new experiences, bring with them changes to weights. On the other extreme, thinkers working in the hard sciences require an abundance of methodologically sound evidence, supporting or undermining theory, in order to adjust the weights that affect what they believe to be true. An ideal for such a scientist might involve the absolute lifting of all trust afforded to anything other than evidence. Even here, truth is always belief, even in the best cases. We ought not be surprised that even the most repeated of experiments -- the requirement for calling a theory a Law -- can fail with increasing scope. So it has been for Newtonian physics: still right, and yet wrong under certain circumstances.

The rest of us, those who make up the bulk of the world's adult population, do not have, and could not have, the benefit of reviewing the methodology or recreating the results of most of the experiments that have added to the body of our civilization's knowledge. We constantly use mental shortcuts to reference and revise the weights we associate with information. Every piece of information we come across, even if its origins are a result of reflecting on oneself, is tempered by conscious or subconscious weighing. Ask yourself, how do you know that gravity is a function of masses and their respective distance to one another? Feynman refers to gravity as the greatest generalization achieved by the human mind. And yet, most of us accept the theory as fact without having gone through the trouble of executing a single test. What hope, then, can the authority on TV expect for his message of the dangers of a given treatment, or some specific threat to our health, if none of us bother with review of method or repetition of experiment? We are merely tourists in the age of enlightenment. Like spies working with vague and incomplete intelligence, we rely on personally assigned, and ever-changing, weights and intuitions for our operative premises. Never being able to rely entirely on evidence, and only seldom referring to it, we rely on our sense of trust.

So, disagreements due to differences in weights are, unsurprisingly, a most common kind of informational disagreement. (The prevalence of cognitive biases is a reflection of how endemic irrational effects on weights are within the human condition.) The resolution to such differences requires convincing an individual to change one or more weights. Those could include the weight they assign to an authority (e.g., the Surgeon General), or the authority that selected him (e.g, the President), or the group with which the authority associates himself (e.g., a political party). More often than not, a change to such weights involves the introduction of some new, or newly understood, context. We often hear of changes to perspective. In its purest form, a change in perspective involves no new information. It merely involves the changing of weights to information already known.

Weights often cascade, as part of the set of mental shortcuts we use to arrive at truth, even though their inheritance may be unsound. With accumulated experience, moving a weight one way or another often requires the movement of an ever greater number of associated weights as well, as our understanding of the world typically ties together explanations of phenomena that could not be revised without adjusting an entire model of reality. More concerning, models of reality are typically shared within a community. It is difficult to avoid damaging the sense of belonging while at the same time holding an orthogonal set of weights regarding socially relevant information. And cosseting our weights from challenge is a small price to pay for the sense of belonging.

Disagreements of interpretation can sometimes be regarded as differences in the assignment of weights. A conductor, when performing a section or phrase of music, may have a preference for a particular tempo. When pressed by students, he may even admit to be willingly disregarding his own interpretation of the composer's intents. But the interpretation of a composer's intent is based on a number of personally assigned weights -- to notes on original copies of scores, remarks in the composer's letters, recollections of contemporaneous musicians, opinions of highly regarded scholars, etc. It may very well be, that it is enough to make available the same information, and spend effort on revising one another's weights, to get two people to agree on an interpretative question. Interpretive disagreements, however, can also be purely aesthetic, which I categorize separately, below. They can also be logical in nature. Disagreements in the interpretation of legal text, for example, can arise as a result of the parties referencing different genera or levels of abstraction for one or more key concepts.

2. Logical

Logical disagreements are disagreements due to the presence of fallacious or unsound argument, or arguments built on such premises. Logical fallacies have been well studied for millennia, and are named and readily located within a taxonomy for ease of reference and communication. One would be tempted to think that this kind of disagreement should be as resolvable as that due to missing information. Surely it should be straightforward merely to point out an instance of hasty generalization or post hoc ergo propter hoc and be done with it. Alas, it is not so. For one thing, logic is associated with intelligence, and people have a tendency to reject out of hand any implication of stupidity, no matter how apparent it may be.

A more concerning issue is the common presence of hidden fallacy. Fallacies can be hidden on purpose, for example, as part of the goals of an ideology, or else by accident. Yet, even these are sometimes difficult to spot.

Of the more common fallacies encountered in day to day disagreements -- cherry-picking, composition, non sequitur, false dilemma -- the fallacy of equivocation might very well rule the roost of confusion. Equivocation causes people to speak past each other without knowing it. As I write this, words such as privilege, value, success, racism, diversity, and many others, are routinely deployed in novel, distorted, watered-down or puffed-up ways, while retaining the cachet or stigma of their more traditional, and distinctly different, meanings.

Equivocation takes a number of forms, and is often hidden in plain sight. Let's take as an example my friend John Sullivan's recent blog post's title, Contrarian Science: The Sun is Conscious. Is it really contrarian? It might seem a step too woo-woo on first blush, but not if one bothers to read the author's careful references to dogma, hubris, assumption, and intuition. It is provocative only to the naturalist, who has a very different working definition of the word conscious, which concerns cellular organisms' sensory and cognitive systems, and their role in promoting organisms' survival via organic reproduction. We can continue down this road, providing definitions for organic, information, cognitive, sensory, reproduction, and so forth. Very quickly, our disagreement, if there is one, disappears, all due to taking the time to define terms. The naturalist simply does not, and cannot, default to classifying the sun as a living organism, even though he may respect others that do so, and even do so himself in moments of poetic license, because doing so as a matter of course would result in an unfathomable degree of absurdity within his field of expertise. Like being the only person at the lab high on ayahuasca, it would be fatuous to expect sustainable, effective collaboration under such conditions.

(It should come as no surprise that contracts routinely go out of their way to define terms. Contract is, as far as I can tell, the epitome of civilization's efforts at broadly achieving justice, and clear definitions are a prerequisite for avoiding contractual misunderstandings. Misunderstandings invariably provoke the sense of inequity, the management and amelioration of which is the purpose and promise of all justice.)

Equivocation comes in many guises. Motte and bailey is a kind of equivocation; its employment begins by making an easily defendable case (the motte), and then extending, by linguistic or conceptual sleight of hand, the argument to a different but similar-sounding concept that is ordinarily far harder to defend (the bailey). Here, the equivocation is not necessarily of a specific word, but of concepts. People familiar with object-oriented languages such as C++ sometimes encounter a related practice, sometimes referred to as a type misuse or type safety violation. This situation arises when a programmer "casts", or treats, an object as an abstraction that is valid theoretically, and allows them to build what seems like a workable program, but which is buggy when executed because the chosen abstraction is invalid for the domain in which it is put to use.

Imagine a child playing with toys scattered on the floor of his room, among them, a set of marbles. A marble is a number of things; we say it inherits from, or is a species of, a number of different categories -- sphere, ball, toy, and also, weapon. We can refer to marbles in all of these ways, theoretically. Functional abstraction of this sort seems to present early in the child's developing rational faculties, as early as the age of 7 according to Piaget. When a parent tells a child to "clean up all your toys" but then encounters some marbles on the floor, the parent's disagreeable state is perfectly grasped by a child who has developed the ability to classify objects in this way. But equivocation is often leveraged purposefully in propagandistic efforts that seek to manipulate language (and, therefore, behavior). Imagine the same situation, but this time, with an ideologue in the role of the parent, one whose frame of reference is such that, anything that could be a weapon, is. "I said, pick up all the weapons!" is meaningless to any innocent child not yet indoctrinated into the ideology. Patriarchy is a term notable for its use in this kind of fallacy, replacing more broadly descriptive terms such as "the powerful" or "the elite" with a narrower, loaded term. Swapping out the abstractions helps support a view of history from the perspective of gender rather than power, or some other, dynamic. Yet, in so doing, it compromises the breadth and predictive efficacy of the theories which rest upon it.

3. Speculative

Perhaps the most common category of disagreement in the political realm is disagreement due to speculation. These are nothing more than disagreements on likelihoods or probabilities of something coming to pass. Set aside for a moment the absence of context to which we referred earlier. Accept for a moment that it is in no way prudent for the political actor to define their purported goal, the timing of achieving the goal, the conditions under which the goal is meant to be obtained, or trade-offs with other unspoken but commonly assumed and desired goals. Even if these were defined by all parties involved in disagreement, what politicians are doing most of the time is speculating on the likelihoods of given outcomes. It surely would be to our benefit if politicians made clear the exact nature of their gambits, for then we could measure the accuracy of their abilities to prophesy, and treat them accordingly.

When we carefully evaluate disagreements of speculation, we notice that, often, the speculative scenario involves or even requires totally hypothetical, and unspoken, premises. We often discover a missing, but required, conditional, if we just parse the argument carefully. What do the advantageous effects of lowering trade barriers rely on? What about the question of opening borders to all comers? Such arguments are often of the form, "by pursuing a certain goal, a Good Thing is likely to result [if only some unspoken condition were true]". Some examples from around the globe:

"Allowing immigrants to enter our nation freely will create a stronger nation [so long as those who arrive are educated, and support rather than destabilize the underpinnings of our teetering economy]."

"Eliminating all tariffs and opening up the markets will benefit all economic actors [so long as our trading partners don't leverage the fruits of free trade to militarily assist an enemy of an ally]."

"Affording named (binned) minority groups public attention and preferential treatment will yield a more morally righteous nation [so long as it doesn't unjustly harm unnamed, ungrouped minorities or individuals]."

Another misleading element of many speculative disagreements concerns those in which the speculation is about an intermediate, rather than final, end. Since the end goal is so seldom spelled out, agreeing on the likelihood of an intermediate end might result in a false sense of agreement. The reader may find it quite probable that shunting young women into STEM specialties is likely to change the gender makeup of certain fields of expertise. But to what end? Bets placed on the ends or purposes of such efforts are seldom spelled out, and when probed, we often learn that the ends are (blindly) expected to be taken for granted as unexceptionally beneficial, or even necessary, by any remotely civilized mind. Such intermediary agreements, in which unspoken dogma assumes an agreement on ends, are typically not worth pursuing, as such agreements are too narrow to convincingly lead to broad agreements on policy or action, and are effectively trivial. They invariably result in unexpected and undesired (even if unintended) consequences, which makes much of the point of agreeing moot.

Oftentimes, speculative disagreements reduce to informational disagreements. There is a tendency to agree on likelihoods of something happening when both parties are in perfect alignment on the facts and their weights (although, as we will see in the next two sections, this is not a necessary result). When disagreeing on likelihoods of outcomes, look out for disagreements on the facts, and especially, the presence or absence of historical precedent. Addressing these may be a more effective approach to changing someone's mind.

4. Preferential

Preferences come in many forms. We are used to the idea of "personal" preferences being aesthetic in nature. But preferences cover so much more than just how colors and styles make us feel. They concern our values as well. Which team we root for, the religious denomination we choose to belong to, the choice of rural versus urban living conditions, the manner in which we communicate, how we learn new things, and what we do in our free time are all examples of behaviors driven by preferences.

It is typical to regard preferential disagreements as common, expected, and seldom worth attempting to resolve. If you are an experienced adult with a preference for traveling in comfort versus style, or for potato over eggplant, what case could your interlocutor possibly have for changing your mind? There is no common ground to which we can appeal when it comes to changing another person's preferences, and preferences are inextricably tied to individuality and authenticity. That's not to say it doesn't happen -- see, for example, the role that marketing plays in modern society. When we attempt to change a free adult's personal preferences, the effort can rightly be viewed as illiberal, due to its inherent disrespect of individual sovereignty. Marketing and advertising must toe this line carefully, by casting their efforts as revelatory rather than manipulative. It's one thing to reveal to someone that they might have been preferring the wrong brand of soda all along, quite another to manipulate them into changing their soda preference. A brightline distinction between the two might not even exist; the ad man is forever walking that fine line, the ideologue doesn't even care to acknowledge its relevance.

Moral disagreements concern either a public normative expression (e.g., guild, religious, or community ethics), or else a personal moral sensibility. Disagreements over public expressions of morally-bound norms are kinds of interpretive disagreements we've already covered. The ethical standards either exist and we disagree on what they mean, or else the public's changing mores require adjustment of ethical standards, which in turn rely on interpreting the public's will, and we disagree on its interpretation. But an individual's moral sense and virtue hierarchy are very hard to differentiate from personal preference. There is no path to convincing someone to choose oathkeeping over pacifism, for example, unless one already knows the individual's moral code well enough to call out relativism or hypocrisy. Individual morality is complex, consisting of fluid hierarchies of value and duty, often positioned in antagonistic, mutually exclusive ways. Cleaving to a prior definition of morally righteous behavior -- moralizing -- is a tell-tale sign of disinterest in arriving at agreement.

When dramatic conflict is moral in nature, it is predicated on the premise that moral matters have no single, universal resolution, known in advance to all. Strictly utilitarian concerns, for example, do not, as a matter of course, override personal duty in instances of high drama. (Here I am referring to classical utilitarianism -- briefly, the moral imperative to maximize the total amount of benefit enjoyed by the world's populace, while minimizing total harms.) In dramatic contexts, strict utilitarianism, or any other moral code or obligation, cannot be a self-evident or irrefutable solution to moral conflict. If that were so, the dramatic effect would be weakened, and the conflict that is requisite to any proper storytelling would be washed out.

Aesthetic and dogmatic preferences have an interesting connection, in that, there might be irresolvable disagreement in the macroscopic question, but a potential for arriving at agreement at the narrower level of "school". Consider the maxim, de gustibus non est disputandum, a Latin expression counseling us on the uselessness in trying to agree on matters of taste. It has a reflection in Picasso's quip, "Ah, good taste! What a dreadful thing! Taste is the enemy of creativeness." Picasso is pointing out the meaninglessness of aesthetic value judgement -- "good" taste as a deleterious preconception. But, as with so many blanket statements, the point's power is diminished and effectively dissolved -- and even repudiated -- when we consider it in narrower contexts. Schools of artistic expression, for example, teach both technique and taste as an indissoluble combination, inseparable for the purposes of teaching the art. An overt focus on one over the other, as is seen in contemporary times with the popular weighing of technique over taste (the athleticization of art) is a signature of corruption within, or rejection of, an artistic school. But for those who have chosen to be students and practitioners of a school, the teaching of taste, and the arrival of consensus on such matters, is not only expected, but allows for establishment of artistic parameters which enhance rather than suppress the creative process. Broad, cross-generational consensus on the outright distastefulness of, for example, overt swaying of the body, or indulgent messa di voce (колбаски), within the Russian classical music tradition, shows how taste can be dogmatic in a preferred, salutary, and teachable way. In so doing, it opens the door to arriving at aesthetic agreement between practitioner and neophyte.

I use the word "school" here to mean a specific way of practice and thought that is sufficiently popular so as to become, through instruction and transmission of knowledge, traditional. For the ancient Greeks, the word hairesis (αἵρεσις) referred not only to a choice of action someone might make, but also to such a school of thought, for example, the Pythagorean, or Eleatic, schools of philosophy. This word eventually evolved into the English word, heresy. Early Christian writers, when using this term to refer to schools of Christian thought, reserved it for those "sects" which they wanted to cast as being outside of the "correct" way of thinking (orthodoxy). As expected, what is orthodox for one is heterodox for another. But within an established school, its orthodoxy can be taught, and disagreements resolved, in a manner indistinguishable from that used by schools of art. Both are deeply traditional, and their conservative nature lends itself to allowing for rightness of preference.

5. Ontological

Finally, there is the possibility of disagreement due to ontological differences. Here I am specifically speaking of differences in our foundational understanding of reality.

Our belief of what constitutes reality, how we weigh the information we come across, and the shortcuts we use in our daily encounters with potential truths or falsehoods are seldom disconnected or arbitrary. They tend to be connected and complimentary in some way. We can think of the way that they are tied together as a worldview. A worldview is more than simply what we think is likely to be true. For those who have formed a mostly coherent set of beliefs, it serves as both a source for why those beliefs exist as they do, as well as a framework for interpreting and integrating new information into an existing set of beliefs. (We might even consider a deeper foundation still, that of a primordial understanding, or "root metaphor", in which a worldview itself may be grounded.)

I categorize disagreement on such grounds separately from those stemming from informational ones, even though ontological differences are quite often a significant cause for informational disagreements. I find it appropriate to distinguish these disagreements into their own category, because disagreements on how we perceive the world can affect more than just our sense of truth or falsehood. Worldviews play a significant role in how our preferences develop, and are likely to affect how we speculate. They can also affect how, and even if, we try to resolve our differences in the first place.

Most people approach reality from some kind of mental foundation to help them make sense of their experiences. Worldviews are often shared within communities, and we can try to derive a given community's worldview from archaeological and anthropological study. But when it comes to resolving a specific disagreement between two people, it can be difficult to evaluate how much of the disagreement stems naturally, and necessarily, from differences in the fundamental way in which the two understand what could be real, or likely to be real, and what could not. This question goes beyond just the weights we associate with information describing reality; it informs us to what the weights could be, and which assertions are within or outside the realm of the possible.

The metaphors people refer to when they analyze and describe their worldviews are some of the most immutable sets of beliefs hosted by the human intellect. They might determine the age one attributes to the universe, the possibility of free will, the manner in which populations are categorized, the approach to how humans treat the natural world, the context for how long a given effort can take before it becomes meaningless, the import of the supernatural, hierarchies of value, and the very point of living. They guide, and sometimes place strict limits on, what we consider to be information, and of course, the weights that accompany any putative information under consideration.

Our foundational metaphors self-select our razors -- those mental shortcuts to which we defer when evaluating information and working out theoretical explanations. Take, for example, the razor of parsimony, which prefers simpler over more complicated explanations, and the razor of preferring a biological explanation over a cultural one. Our worldview can determine if and when either razor is in effect, and, importantly, whether the two are related. The fundamental building blocks that form my worldview, for example, naturally lead to my heavy reliance on evidence for my ontological outlook, and select probabilities based on track record. My understanding of reality comfortably accepts Occam's Razor as a reliable best practice, and this has a correspondence in my acceptance of the razor which, in questions of human behavior, prefers the biological explanation. I see the latter as a mere extension of the former. And yet, preferring a biological explanation over a cultural one is certainly not something I can expect of my friends and intellectual sparring partners. Indeed, it is the very source of my most profound disagreements.

Worldviews, after all, act as lenses through which we observe and understand reality, and the question of whether it is even possible to exist free of their distortions is itself dependent on one's worldview. It doesn't help matters when worldviews are incoherent or underdeveloped, which is the case more often than not. And it is impossible to avoid or ignore one another's worldviews when trying to communicate sensibly. Even this essay is predicated on the author's worldview -- one the reader may not share. Importantly, individuals holding differentiated worldviews need not disagree on all conclusions, and are often found to agree on approaches to knowledge. I am not a religious person, but many of my dearest friends and mentors are. It should seem amazing that we can agree on as much as we do, when we might not even agree on the relationship between the Right and the Good. This is because the resolutions to many important, practical disagreements do not hinge on agreement to such questions. The ability to agree on a number of truly broad questions might only require an understanding and acceptance of other worldviews, not necessarily adjustments to our own. Still, when true ontologically-based disagreements arise, they can be intractable.

Seldom can we hope to change one another's worldviews. What we can do, however, is help one another work out the specific metaphors that lie at the base of how we understand reality, and provide guidance when those worldviews clash with expressed informational weights. In other words, our hope of coming to agreement rests in part on our abilities to help others form a sufficiently comprehensive worldview, such that disparate beliefs, grounding information, and conclusions hang together coherently for the individuals themselves. This exercise is always worth doing, even if, in the end, we must agree to disagree.

- A. Yampolsky, Los Angeles, 2023