Metaethical Naturalism
1. Introduction to Metaethics
Morality consists of distinctive ways of influencing the feelings, thoughts, behaviors, actions and decisions of people. Moral reasons seem to appeal to a different kind of authority than strictly self-interested reasons. Although people may differ in the contents of their morality, it seems that there are common presuppositions to different substantive moral views. It could be that all moralities appeal to the same kind of authority or motivational force (God and hell, say), despite differences in the content of their prescriptions. "Metaethics is the attempt to understand the metaphysical, epistemological, semantic, and psychological, presuppositions and commitments of moral thought, talk, and practice."1 Metaethics studies the common denominators of everything 'distinctively moral', so different substantive ethical positions and their disagreements do not perturb this task. Indeed, they contribute to it by serving 'data' to the metaethicists. Normative ethics tries to accurately describe moral phenomena (or moral phenomenology, the two shall be used interchangeably). They try to formulate principles that would generate prescriptions that accord with our actual moral intuitions, thoughts and practices. Of course, ethics is not strictly descriptive. Its very practice also contributes to shaping our moral intuitions, thoughts and practices. Applied ethics is concerned with finding moral answers to concrete situations, in other words, generating prescriptions. So ethics also changes and creates moral phenomena, whereas metaethics strictly studies moral phenomena as they are.
Ethical phenomena are studied by metaethicists in a manner analogous to natural scientists. Natural scientists study natural phenomena which are not dependent on humans, such as inanimate objects, other animals, the planets. They are trying to understand the 'powers behind' these observations, to use a Lockean term. They cannot change the manifestations of these phenomena, as they are not dependent on humans. All they can do is study how they actually are. In the same way, metaethicists try to study moral phenomena. The difference is, this group of phenomena is dependent on humans. Humans contribute to changing and creating moral phenomena. This human-dependency blurs the line between the observer (the metaethicist) and the observed. In practice, this blurring can make metaethics seem like a discipline completely different from natural sciences. But in principle, metaethicists are trying to do the same thing as natural scientists, just with moral phenomena. They are trying to accurately understand a group of phenomena, and to this end are required to come up with a theory that is compatible with moral phenomenology.
Phenomenological fit does not exhaust a metaethical account’s requirements, however. It must also be elucidating and plausible with regards to philosophical considerations of ontology, epistemology, semantics and psychology, among other branches. A sufficiently complete explanation of any single metaethical theory requires positioning it along such dimensions. Due to this multifactorial complexity, I will focus my explanation of metaethical naturalism on a single version of it, that of Railton's synthetic naturalism. Moral ontology asks whether moral properties or facts, independent of human judgments and beliefs, exist; if they do, what their nature is. Moral realism is the ontological claim that such independent moral properties or facts exist. Moral epistemology asks whether we can attain moral knowledge and how. Moral semantics asks whether moral judgments are truth-apt, i.e. whether they can be true or false. Cognitivism is the claim that moral judgments are truth-apt, because they are descriptions. Cognitivism is also the psychological claim that moral judgments are cognitive states I.e. beliefs, because beliefs are identified with descriptions. Psychological non-cognitivism is the claim that moral judgments are non-belief states, such as emotions, which are not truth-apt. Metaethicists have not yet been able to show that there is only one correct order of inquiry. There is no objective standard for what realm of inquiry should be settled first. The same question can be tackled from two different starting points. For example, the question of whether moral judgments are cognitive or non-cognitive can begin with semantics, psychology or ontology. The same substantive position can be arrived at from many starting points. The upshot is that in choosing which subdiscipline to begin with, one is already making an assumption. One's choice of the first step into metaethics reveals one's most fundamentally held presupposition.
2. Motivations for Naturalism
So what is the most fundamental presupposition of naturalists? More than anything else, naturalists take natural phenomena for granted. That is, the only things they consider to be real are observable things, and the properties and facts that explain them. Accordingly, naturalists think knowledge is acquired empirically. So the primary commitments of naturalists are ontological and epistemological. Their metaethical starting point reflects this. If ontological-epistemological naturalism is true, then there is no prima facie reason why naturalism should not also apply to morality. Call this the argument from coherence or consistency. It seems incoherent to be naturalistic towards everything except morality. What does it mean to apply naturalism to morality? It means that moral phenomena are treated like other natural phenomena. The empirical method that applies to studying natural phenomena also applies to studying moral phenomena. This empirical method is exactly a distinctive characteristic of naturalism a metaethical position. Observations are made, and hypotheses are formulated in attempt to explain these data. Moral phenomenology is the data for metaethical theorizing, as mentioned. Though not the only requirement, any metaethical theory must be compatible with moral phenomenology. But the same phenomena can be explained by many different theories. So simply referring to seemingly supportive phenomenology isn't a good enough argument for a position. The adequacy of phenomenological evidence depends on whether a given metaethical position gives not only plausible, but the only plausible explanation for the phenomena. Railton claims that the only or best explanation for moral phenomena is a naturalistic explanation, i.e. one in terms of only natural properties or facts. It explains moral terms without referring to moral terms and is thus reductive.
A feature distinguishing Railton from some other versions of metaethical naturalism is his 'synthetic', i.e. a posteriori method. On this method, not only the directly observable exist. Something also exists if it adequately explains phenomena. So if the naturalistic explanation indeed adequately explains moral phenomena, then the natural properties or facts that feature in it would exist i.e. naturalism would be true. What is the reasoning behind this idea, that the properties and facts featuring in an explanation themselves exist? As mentioned, naturalism starts by taking for granted that phenomena are real: that is real which we can directly perceive. The properties or facts in explanations are not directly observable, yet they are still posited as real. Here Railton specifies another way in which something can be real: That is real which we can interact with and be causally impacted by. This possibility for informative and causal interaction is called the 'feedback' feature. It is one of Railton's criteria for reality: A property or fact must be able to give us feedback to be existent. What sort of non-perceivable thing might be able to give us feedback? The intuitive answer is: something that 'regulates' phenomena, because then it can give us feedback indirectly via the directly observable. Phenomena manifest in a certain way, and the 'powers behind' such manifestations can plausibly be said to exist, even though they themselves are not directly observable. They explain why phenomena behave as they do; they cause them. These 'powers behind' are the properties and facts posited in explanations. If powers indeed exist behind the phenomena, then the properties and facts that stand for them also exist.
When can such powers be said to exist? Here we invoke the fact that we experience natural phenomena as very consistent, coherent, stable, i.e. patterns. In interacting with natural phenomena, we can manipulate them to a certain degree, but the patterns are independent of us. There must exist some powers independent of us that cause these patterns. The existence of natural properties and facts is a plausible explanation for phenomenal patterns’ independence from us. This is Railton's second criterion for 'reality': A property or fact must be independent from our subjectivity to be existent. These phenomenal patterns persist regardless of our subjective desires or beliefs, so the properties and facts that explain them can be said to exist. Positing the existence of independently real properties or facts is not justified by mere intuitiveness. We deny it at odds with our experiences: it is clear that we experience certain patterns as outside our control. "For although our conceptual scheme mediates even our most basic perceptual experiences, an experience-transcendent reality has ways of making itself felt without the permission of our conceptual scheme- causally."2 In summary, if moral phenomena are consistent, coherent, stable and independent of us, then by the same rationale as for natural phenomena, the properties and facts that feature in the best explanation of moral phenomena can be said to exist.
Explanatory power is usually measured as descriptive and predictive accuracy, and this requires falsifiability via empirical tests and comparisons. Falsifiability is also considered an advantage by Railton. Like Popper, Railton thinks falsifiability is a virtue in a theory, and considers it a methodological advantage of the empirical method. To arrive at his theory, Railton uses a method similar to that of observational studies in science.3 Certain phenomena are identified as relevant to morality. Then, a naturalistic theory is formulated to explain these observations. The theory implies certain expected observations, which are compared with actual observations. If they do not match, the hypothesis must be rejected or revised, so it is a trial-and-error process. Finally, the theory must be compared with alternative theories, to ascertain that it explains the phenomena better than these alternatives. We will apply these four step to each of the crucial notions in Railton's metaethical theory.
3. Nonmoral Good
Fact: Objective Interests
Railton takes moral phenomena to consist of facts about human beings from the natural and social sciences such as biology, psychology, history, sociology, economics… “We are natural and social creatures, and I know of nowhere else to look for ethics than in this rich conjunction of facts."4 The most fundamental facts are biological: All living organisms, including humans, aim to survive and thrive. There are facts about how humans are biologically constituted, thus facts about human needs i.e. conditions that must be satisfied in order for humans to survive and thrive. Importantly, such conditions do not depend on anyone's beliefs or preferences: they are the objective needs of humans. For example, we can think whatever we want, but this won't change the fact that we need to drink water at least once every three days. Railton calls these objective conditions for the surviving and thriving of humans their 'objective interests'. Besides facts about humans themselves, there are facts about their existential conditions, for example their environment. These conjunction of facts yield facts about how effective different human actions, choices, behaviors and strategies are at satisfying human needs. Which substantive moral theory is correct depends on all these facts about human objective interests, existential conditions and alternative strategies.
Subjective beliefs and desires are secondary qualities, because they depend on, supervene a 'reduction basis'. This reduction basis consists of primary qualities like a person's constitution and circumstances, in other words, the same things that constitute a person's objective interests. So a person's subjective interests and desires supervene on the factual conditions that make up their objective interests. But this does not mean that a person's subjective interests necessarily reflects their objective interests. People can believe or want things that are not in their objective interests! It is true that human interests exist at all because humans exist, and in this way the existence of objective human interests is dependent on the existence of humans. But if humans exist, then the facts about human interests are independent of humans' subjective beliefs and preferences. The independence i.e. objectivity of these interests mean that all humans have in-principle reasons to be motivated by and comply with them, regardless of differing beliefs and preferences. This is precisely what gives authority to a morality based in objective values. It is what guarantees a certain degree of compliance from every normal person (meaning one whose motivations do not deviate from the average organism's desire to survive and thrive). The authority of morality derives from it being strategic to comply with if one wants to satisfy one's objective interests, some of which are universal to humans. On Railton's account, objective human needs are what gave birth to morality, what sustains it, gives it authority and ensures compliance with it. We shall see later that moral precepts may be conventionally adopted. But their authority is derived from convention-independent values. It is this objectivity that helps naturalism sidestep relativism. “[T]o whatever extent specific moral rules and principles are products of convention, their claim to authority relies on some standards that are not products of convention.”5 It follows that judgments about what is in our objective interests are truth-apt— they can be true or false.
Theory: Nonmoral Good
From the facts that there exist objective interests, subjective interests, and that the two do not always agree, Railton begins his theorizing. He introduces the notion of "objectified subjective interest'. A person's subjective beliefs and wants can align more or less with their objective interests. When the two align, Railton calls this objectified subjective interests. Railton suggests that objective interests can be found by conceiving the desires and interests of a person in an idealized state, i.e. with full knowledge of circumstances, full self-knowledge, full reasoning capacity, mental and emotional clarity for rational decision-making... Objective interests are what such a perfect version of a person would want for the imperfect version of themselves that they are. The idea is that an omniscient, omnipotent, self-interested version of you would know what is actually good for you. This idea would explain why many people fail to know or want what is actually good for them: they are in fact not omniscient, omnipotent, perfectly rational and emotionally calm… So objective interests are neither essentially nor categorically motivating (so Railton is an externalist of motivation), nor are they necessarily knowable.
Railton then defines an individual's nonmoral good in terms of this idea of objective interest: “X is non-morally good for A if and only if X would satisfy an objective interest of A.”6
He also extends this definition to define an intrinsic nonmoral good. The addition is simply that the objective interest is good in itself without reference to other objective interests. The value of an intrinsic good basic i.e. it is not derivative of other goods. Both definitions are naturalistic because they are in terms of objective interests, which consist of naturalistic properties and facts, as previously mentioned. They are Railton's moral 'hypotheses'. They must be shown to explain i.e. accurately describe and predict moral phenomena. So let us derive the observations to be expected, if the hypothesis of nonmoral good is true, and compare them with actual observations.
Expectations
What is nonmorally good for an individual is basically what is conducive to the survival and thriving of their organism and species. Add to this the fact that emotions are what move us, what motivate us, towards or away from some goal or means. Taking the two together, it is likely that humans would have evolved the trait to be motivated by and feel positively towards their objective interests, because such individuals would have survived more often than individuals who were averse to pursuing their objective interests. The former's descendants, current humans, would have inherited these emotional and motivational dispositions. So we expect to observe that current human beings tend to feel positively towards and be motivated by their objective interests. That is, we expect to observe that people will be happier and more satisfied, the more their objective interests are satisfied, and the more their subjective desires and beliefs align with their nonmoral good (consequently that they pursue their nonmoral good more often). Actual observations seem to confirm these expectations: Social studies show that people who have their basic objective interests covered, i.e. sustenance, housing, security, are happier than people who don't have such basic needs covered and constantly worry about them. Addicts are people who subjectively desire things that aren't good for them, and they seem on average to be less happy than non-addicts. Addictions are an example of desires contrary to what is nonmorally good for a human, and this is a phenomenon expected by Railton's defining nonmoral good as independent of subjective desires. So people's motivations often, but not always, align with their nonmoral good. These observations bear out Railton’s claim that nonmoral good is not essentially or categorically motivating.
In this connection, an important observation is that when people's beliefs about what is good for them changes, their motivations also change. So motivation also depends on beliefs about nonmoral good, not nonmoral good alone. But nonmoral good is also independent of subjective beliefs. What is nonmorally good for us is a matter of fact, not subjective belief. Judgments about nonmoral good are truth-apt. This implies that we can expect to observe instances of people having false beliefs about what is actually good for them, which in turn motivates them away from their nonmoral good. This expectation is confirmed by observation. For example, people can be mistaken about what a healthy diet consists in, and are consequently motivated to follow a diet that perpetuates their disease. So in many cases, people are motivated to pursue what they believe is nonmorally good for them, not necessarily what is nonmorally good for them. Subjective interests do often align (approximately) with nonmoral good, however, and nonmoral good is determined by the reduction basis (the natural properties and facts of the person and situation). SO people with similar reduction bases can be expected to have similar subjective interests. This expectation is confirmed by observation: Subjective interests arez most similar in areas of life where people's objective interests most overlap i.e. basic needs such as food, shelter, security… Similar people tend to share preferences or patterns of behavior and decision-making, e.g. people who share similar constitutions i.e. personality, health, sex or similar situations i.e. background, environment, socioeconomic status, culture... For example, physically gifted people often become athletes, technically gifted people become engineers, agreeable people go into careers that focus on nurturing others...
Evaluation
These are a few examples of observations that confirm Railton's hypothesis i.e. naturalistic definition of nonmoral good. Since nonmoral good accurately describes and predicts them, we can say that the natural property of nonmoral good explains moral phenomena well and that it therefore exists. Here Railton points out a potential objection to his hypothesis: The property of nonmoral good is constituted by a reduction basis. Isn't the notion of nonmoral good redundant then? Isn't the explanatory work actually being carried out by these other, more basic natural properties and facts of the reduction basis? To this Railton responds that an explanation is not made less useful by the fact that it is reducible to more basic phenomena. Chemistry is reducible to Physics, and chemical notions reducible to physical notions. But this doesn't make Chemistry any less useful and elucidating, and it doesn't make us think of Chemistry as any less real. We consider chemical properties and facts real, because they indeed describe and predict phenomena reliably, just on a level of analysis higher than physics. You could say chemical explanations abbreviate physical explanations so that we can get more work done. Chemical terms consist of clusters of physical phenomena, and are used instead so that one doesn't need to go through these longer lists of constituent physical phenomena every time. In the same way, moral terms, for example Railton's nonmoral goodness, consist of more basic natural properties and facts, but are still useful and real.
Railton then points out an advantage of his hypothesis over alternatives: Moral phenomena cannot be exhaustively explained by subjective interests alone. Examples were given above where people could be disadvantaged by their own subjective values, beliefs and desires, where they 'ought' to do otherwise than their subjective interests would tell them. Any theory that lacks an objective standard for what is in people's interests (such as Railton's nonmoral good) gives an incomplete and superficial explanation of social phenomena. Even more importantly, a lack of a convention-independent standard of value leaves morality without authority, as we have mentioned. An objective, independent standard of value generates reasons to act that are authoritative and compelling regardless of subjective desires and beliefs, what are called normative facts. "Thus emerges the possibility of saying that facts exist about what individuals have reason to do, facts that may be substantially independent of, and more normatively compelling than, an agent’s occurrent conception of his reasons."7 Normative facts are not necessarily motivating, however i.e. Railton is an externalist of motivation. For example, facts about what is objectively good for you are not motivating if you do not believe them. But if you did believe them, you would be motivated by them (though not necessarily overridingly). Normative facts are in-principle, or ‘hypothetically necessarily’, motivating. This brings us to an important objection: the epistemic is-ought distinction. It claims non-cognitivism, that is, that moral judgments are not cognitive like factual or logical judgments. This means that moral judgments are neither descriptions nor truth-apt. Railton claims that there are moral judgements that are indeed facts, namely the normative facts, so he must refute this argument for non-cognitivism. He must show that there is no such difference in epistemic kind which prevents value judgments from being factual statements, that both are cognitive. Let's see how Railton deals with two of the most important arguments for an epistemic is-ought distinction.
The first objection claims that factual and logical judgments are characterized by instrumental rationality, without any presupposed ends or a value system. By contrast, value judgments do presuppose ends or a value system. Their cite as evidence the observation that two equally rational people can disagree about values. This implies that rational arguments do not apply to disagreements about value. They cannot decide disagreements of value, because the problem lies not in a difference of instrumental means, but a difference in presupposed ends. No amount of factual or logical arguments can put a dent in a difference of values. Railton responds that this argument is treating values, or presupposed ends, like subjective preferences without basis in any objective standard. But as we have seen on Railton's account, subjective interests supervene on objective interests, and the latter are a matter of fact, not preference. Disagreements can then be reduced to a common basis of normative facts, which consist both of facts about what is instrumentally rational to do (which this epistemic is-ought argument concedes are common to disagreeing parties), and facts about which values are rational to pursue. This second class of facts are the value judgments in question. Value judgments are still facts, the subset of facts about what can in-principle motivate humans, given that all human (for the most part) share the goal of surviving and thriving, and that all humans are innately motivated to be rational in the extended sense. Value judgments are facts about which subjective interests have high probability of approximating nonmoral good- they are predictive or hypothetical facts, rather than actual facts.
The second objection comes from Hume. Hume claimed that moral facts, if they existed, would be essentially practical i.e. motivating (internalism of motivation), though not necessarily overridingly motivating. By contrast, Hume claimed that nonmoral facts are not essentially motivating. Therefore, moral and nonmoral facts cannot be of the same kind. Railton, being an externalist of motivation, simply rejects the claim that moral facts i.e. normative facts would necessarily be motivating if they existed. And his account, as we have presented it until now, has attempted to show that normative facts need not be essentially motivating to retain their authority. People have reason to pursue or do what is rational, whether they subjectively want to in their current state or not. Again, moral facts are facts about what would in-principle motivate people. In conclusion, there is no epistemic is-ought distinction, Railton argues. Moral judgments are cognitive and are subset of factual judgments.
4. Individual Rationality
We have only talked about what is factually in someone's interests so far. What about what humans ‘ought’ to do? As seen, 'ought' statements are a subset of 'is' statements for Railton. 'Ought' statements are merely explanations of the instrumentally rational thing to do in contexts with presupposed ends. Both the presupposed ends and the instrumental means can be expressed in terms of natural properties or facts. Hence, 'ought' statements can also be formulated as 'is' statements. Railton calls 'ought' statements criterial explanations, because they explain something in relation to some criterion, the presupposed end. Their factual nature and explanatory function is clear by the fact that ought statements can be expressed as conditional statements. For example, 'Houses ought to be well isolated' is an explanation relative to the criterion of 'keeping warm', a naturalistic criterion. This ought statement abbreviates the conditional statement: "If one wants to keep warm, then the instrumentally rational thing to do is to build well-isolated houses." Humans 'ought' to do what is nonmorally good for them (if they want to survive and thrive). They should choose means that are conducive to their nonmoral good. How do they do this? This is where Railton introduces a second important biological fact: humans have feedback mechanisms.
Fact: Feedback mechanisms
"Humans are creatures motivated primarily by wants rather than instincts."8 And they tend to want things that feel pleasurable, and avoid things that feel painful. This is a feedback mechanism, and it is not a rationally guided process. It is merely a process of stimulus-response conditioning. The more immediate and intense the stimulus, the stronger a certain response is conditioned into a person. The more immediate the effect, the more it is associated with what immediately precedes it, a response to which is conditioned. As mentioned, evolution would have ensured that humans were motivated to pursue what was conducive to our survival and thriving. It did this by making what is objectively good for us pleasurable, and what is not objectively good for us painful. So by the aforementioned feedback mechanism, we tend to repeatedly choose or do things that are objectively good for us, and tend to stop choosing or doing things that are objectively bad for us. Over time, by trial and error, our desires and motivations get updated and adjusted to better fit with our objective interests.
Feedback mechanisms guide our motivations towards our objective interests; they make us sensitive to what is objectively good for us. So they are one of the ways in which we have epistemological access to our nonmoral good. Without them, we would be aiming at achievements that either make no difference or outright leads to our downfall. But feedback mechanisms are also fallible. They function on the basis of feelings of pleasure and pain. But such feelings are not always in accordance with our objective interests. Furthermore, they are very programmable, by factors such as habituation, upbringing, culture, social pressure… The consequence of this programmability is that humans can be trained to want many different things, including things that are bad for them. Flexibility isn't always a good thing, in gives more room to being misled. Another very clear disadvantage of this, and evolved traits in general, is the subordination of long-term needs to short-term needs. Evolution would have always favored actions and choices that guaranteed the immediate survival of an organism, potentially at the cost of longer term thriving— there is no future without a present. For example, organisms only needed to survive long enough to reproduce. What happens after successful reproduction is of no concern to evolution. This produced the human propensity towards instant gratification. But what is instantly gratifying may not be objectively good for us in the long run. Typical examples of this are health vices such as smoking, drinking, overeating...
Theory: Extended Individual Rationality
So humans 'ought' to pursue what is in their objective interests, and they most often do this nondeliberately via feedback mechanisms. Feedback mechanisms work for the most part, and allow humans to increasingly pursue and satisfy their objective interests throughout life. In other words, feedback mechanisms make humans increasingly efficient at attaining their objective interests. Efficiency in achieving nonmoral good involves two distinct and equally necessary tasks: aiming at the right ends, and choosing the most instrumentally rational means to these ends. The person must both want what is actually good for them, and choose the most efficient means to these ends. The feedback mechanism works on both parts. Through feedback, humans adjust both their wants and behaviors i.e. means. This is where Railton introduces his notion of Extended Individual Rationality, from now on called individual rationality for short. It is the dual human capacity to aim at objectively good ends, and choose the most efficient means to these ends. A complete form of rationality must include both a capacities. Individual rationality can be implemented by feedback mechanisms. This is one of its nondeliberate instantiations. But individual rationality can also be implemented by conscious deliberation. So the concept of individual rationality phenomenally exceeds feedback mechanisms. Not only that, its explanatory power exceeds that of feedback mechanisms. The feedback mechanism only tells us that a certain behavior has been reinforced or discouraged because it felt pleasurable or painful. The notion of individual rationality adds that feedback mechanisms produced certain results and not others, because the former are conducive to the individual's nonmoral good. That is, individual rationality is a criterial explanation, because it refers to a presupposed end, that of humans wanting to survive and thrive. Individual rationality, defined as efficiency in achieving nonmoral good, is a naturalized criterion. A naturalized criterion is empirically verifiable, and it doesn't get much more empirical than an ‘engineering’ concept like efficiency. Given the specified goal of biological survival and thriving, it is empirically verifiable whether or not this goal is achieved, and whether the means used more or less resources, time...
Fact: Motivational System
Not only does individual rationality have explanatory power, Railton claims that as a human capacity, rationality is also intrinsically motivating. People are concerned about being rational. Rationality has this pull on people in and of itself, not based on explanatory power. This is due to a third biological fact: both deliberate and nondeliberate mechanisms for approximating rationality are anchored in the human motivational system. This makes sense from an evolutionary perspective- it is empirical evidence confirming our earlier argument that evolution would have programmed into people emotional and motivational dispositions conducive to the satisfaction of their objective interests i.e. rational dispositions. In other words, evolution would have ensured a systemic connection between motivation and rationality. To summarize, individual rationality is both explanatory and normative. Importantly, its explanatory power does not necessarily imply its normative force. A notion can be explanatory without also having normative force. It is an additional feature distinct to Railton's account, that individual rationality should also has normative force and this due to its being based in the human motivational system. This is how individual rationality goes from being naturalistically defined as a strictly explanatory criterion, to being a normative criterion which genuinely motivates. What is initially purely descriptive becomes regulative, given that the context of human life presupposes a nonmoral good.
Theory: Normative Force
We have now accounted for the source of the normative force of rationality, namely its factual anchoring in the motivational system. This is where Railton extrapolates and claims that the normative force of rationality is that which gives normative force to normative facts in general. This is his third hypothesis, that rationality's anchoring in the motivational system is the source of all normative force! Any normative fact, if it indeed has normative force, derives it from the innate human motivation to be rational. We previously mentioned that the existence of an objective standard of value, i.e. nonmoral good, gave rise to normative facts, facts about what individuals have reason to do that are independent of their subjective desires and beliefs. There, the normative facts enabled by nonmoral good were restricted to facts about which ends are rational to pursue. Extended individual rationality expands this realm of normative facts to include facts about instrumentally rational means. So normative facts include both facts about rational ends and rational means. Being thus based in rationality, normative facts can leverage the normative force of rationality. This is Railton's reasoning for his theory, that normative facts derive their normative force from humans’ innate motivation to be rational in the extended sense.
Expectations
What observed phenomena can be explained in terms of individual rationality? Individual rationality can explain all learning. Why did the child eventually learn to delay gratification? Because it pays off in the long run i.e. it was more rational. How did the child do this? By the human feedback mechanism, which aims at rationality and optimizes for it. Individual rationality can explain why people's goals and preferences change over time. Many people say that their priorities changed as they got older. They became wiser, more rational in their desires and motivations. For the most part, we can expect that people tend to become more efficient as they gain experience. This can be because they gain knowledge with which to make increasingly rational deliberations. Or it can be that experience has conditioned them towards more rational behavior unconsciously, through the feedback mechanism. We can expect that rationality is most fostered in areas where feedback is clear and immediate. We can expect that the more rational a person is (i.e. the more they can control their irrational desires and the better judgment they have about means), the more they will satisfy their objective interests i.e. the more satisfied they will be. We can expect that what is good for someone is best known by that person themselves. This is because the feedback mechanism teaches individuals about their own needs and what works for them. It does not teach them what others need or what works for others. If individual rationality is indeed normative and in-principle motivating for humans, what observations can we expect? We can expect that people most often try to be as rational as they can be. They will be motivated by what they believe to be rational. The innate drive to be rational can explain why people upon deliberation usually choose the more rational option. They aim at this because they (most often) want what is best for themselves, and believe this is what they need. All of these attempts are of course potentially hindered by other human traits such as contrary emotions, dysfunctional regulatory systems... But this preference for rationality implies that rational choices tend to win out; rationality tends to be reinforced in a person and become dominant over time. All these expectations about the optimal conditions for rationality and its innate motivating force are, according to Railton's preliminary analysis, borne out by actual observations.
Evaluation
The existence of rationality isn’t usually doubted, but the existence of distinctly normative motivation is. Railton presents the fact that the human capacity for rationality is innately motivating, due to its anchoring in the motivational system. The advantage of using rationality in this way is that rationality’s existence can be used to prove the existence of a source of motivation that is independent of subjective desires. Any moral skeptic who believes in the existence of human rationality, but not in the normative force of that selfsame rationality, is simply empirically mistaken. There is another very important implication: Normative facts based in individual rationality can bridge the empirical-normative gap. Normative facts are empirically based in the objective interests of humans: everything conducive to survival and thriving is valued by humans, and this is a biological fact, not some invented metaphysical value. In turn, sensitivity to objective interests and instrumentally rational means to them have been built into humans via feedback mechanisms and deliberate rationality. An innate motivation to accord with these facts was also built into the human motivational system. So normative facts, epistemological access to them and motivation to accord with them are all grounded in human biology and environment. Values, both primary i.e. nonmoral good, and derivative i.e. individual rationality, have a factual basis in human biology and environment. In other words, there is no ontological distinction between values and facts; both are based in natural phenomena.
5. Social rationality
Fact: Morality's scope
So far, we have accounted for individuals; their objective interests, the capacities which enable them to learn about and approximate these (feedback mechanisms and rationality), and derivatively, facts about what can in-principle motivate them. We have thus accounted for individual normativity, what can give individuals reasons to act that are independent of their subjective desires and beliefs. This is an indispensable element of morality, that moral concerns be authoritative, compelling and compliance-inducing despite competing subjective concerns. But normativity alone is not sufficient for values or norms to count as distinctively moral. One additional element is required, namely the social element. The scope of morality is social interaction. The problem of morality is to find out how people should interact with and behave towards one another, either when living in close proximity or when actively associating with one another. That the social is the scope of morality is not a natural fact per se. It is merely convention that we call social prescriptions and their study 'morality'. But it is a natural, convention-independent fact that social phenomena exist, that humans inevitably interact with each other and other species. The possibility of studying this group of phenomena, then, is also convention-independent. Whether or not it is called morality is irrelevant. Morality is simply the name given to the study of social phenomena, more specifically the study of systems for regulating social behavior i.e. social prescriptions. The task of any morality is to come up with a set of norms, prescriptions, rules for social interaction and behavior.
Theory: Moral Rightness
Just as there are normative facts for individuals, so morality tries to establish social normative facts, facts which give participants to an interaction reasons to act in a certain way, reasons to comply with certain social norms, that are independent of their subjective desires and beliefs. “We thus may say that moral norms reflect a certain kind of rationality, rationality not from the point of view of any particular individual, but from what might be called a social point of view.”9 Developing this idea, Railton specifies social rationality using the components of individual rationality, just applied to a group instead of only one individual. The hope is that the normative force of individual rationality can be lent to social rationality, so that social norms can be as authoritative and motivational as normative facts about individuals. This is the beginning of his hypothesizing about morality, because he makes the assumption that facts about individuals can apply to groups in this way, that something said of an individual can be extrapolated to a group, of course with qualifications. He also cites factual support for such assumptions and qualifications.
Social nonmoral good
Just as there is an objective nonmoral good for an individual, so there is an objective nonmoral good for a group of individuals. The objective good of a group is simply the aggregation of the individual members' nonmoral goods, with the qualification that individuals' interests are not completely independent from each other. There are interrelations and interdependencies between different individuals' interests. For example, humans would not only have evolved traits that benefited us individually, but also traits that benefited the human species as a collective. We would not have evolved as strictly self-interested, but also as caring towards our offspring, say, even though they cannot 'repay' us. Individuals who lacked inclination to cooperate and reciprocate might have been eliminated by the tribe, so that those who remained, and us their descendants, have more empathy. The upshot is that the aggregation of individuals' nonmoral goods would not amount to exactly the sum of each individual’s separate interests. But social good is nonetheless the taking into consideration of every individual's nonmoral good. As evidence for this assumption that the social nonmoral good is an aggregation of individuals' goods, Railton cites the historical trend that most of the different moralities from different cultures, throughout different times, have connected their norms, their codes of conduct, to human interests. Railton calls this the humanization trend. The various moralities, religious or not, advertised for their codes of conduct by claiming many benefits to their followers' interests, importantly, to individuals' interests. They promised individuals life after death, peace of mind, riches, the preserving of their possessions, luck and talent to their children...
Impartiality
Like individual nonmoral good, social nonmoral good is also idealized. It is what we would know to be good from a social point of view, if we had full knowledge, were perfectly rational in our deliberations etc. Another idealized characteristic of social nonmoral good, not present in individual nonmoral good, is that of impartiality. “[Social rationality is] what would be rationally approved of were the [objective] interests of all potentially affected individuals counted equally under circumstances of full and vivid information”10 It is very important to note that this requirement of impartiality does not rest on a presupposition, such as modern ideas about human rights, intrinsic value of individuals, liberalism... Such an assumption would be inconsistent with the naturalist framework. The requirement of impartiality derives from the existential fact that nature does not discriminate: Power can be possessed by anyone, and anyone who can fight for their own interests, will. Assuming that most parties in an interaction are capable of defending their own interests to a certain degree, the result is that everyone's interests are accounted for i.e. impartiality. Let us elaborate on this fact of life: Not only do humans inevitably interact, but they inevitably interact competitively. It is a biological fact that living organisms compete for resources, territory... Each species does not compete against all other species, only those with whom they share needs, leading to conflicts of interests. For example, plants only compete with other plant species that require the same nutrients. Species that require different nutrients can coexist and even form symbiotic relationships. Accordingly, humans have historically competed with one another in cases of conflicts of interest. Any party who was strong enough to defend and fight for their interests did so, thus imposing their demands upon others. Those who were strong enough just got their way completely by coercing others. Those weaker could have banded together to become stronger. Where the power balance is roughly equal, parties to the conflicts of interest negotiate with each other. Importantly, negotiations begin out of necessity, not out of fine moral feelings such as social justice or universal respect for rational autonomous agents. If the parties did not negotiate and reach an agreement that sufficiently satisfied all parties' interests, the war would go on where each party fought for their own interests by force rather than words. With this human existential condition in view, we specify the task of morality even further: to generate norms for social interaction and behavior that sufficiently satisfy the interests of the individuals or parties in interaction. Sufficiency is achieved when none of the parties to the interaction are motivated to instigate reforms of the norms. This requirement has an implication: In order to satisfy every individual in a society, the norms must sufficiently satisfy each and every individual’s nonmoral good. This is another piece of factual support for the hypothesis that social nonmoral good is an aggregate of individuals’ nonmoral goods.
Social Rationality
It follows that to satisfy individuals' nonmoral goods, the adopted norms must allow instrumentally rational means to these goods. So the second component of social rationality, just like in individual rationality, is instrumental rationality. Individuals can be instrumentally rational either deliberately, by conscious thinking and decision-making, or nondeliberately, by feedback mechanisms. Railton thinks that societies can be instrumentally rational in the same ways. People can make rational deliberations in a group e.g. government. And it is reasonable to suppose that feedback mechanisms also exist in societies, however with the qualification that social feedback would be less direct, immediate and reliable than individual feedback, since there are so many more variables, some potentially confounding. In individuals, both the feedback is more direct and clear, and the following attempted interventions apply more directly and effectively. Social feedback mechanisms would thus be less effective than individual feedback mechanisms. Railton cites historical evidence for the existence of social feedback mechanisms. The sensitivity to losses of entitlements, or loss aversion, is one such social feedback mechanism: "A common theme in the [historical] works cited in note 32 is that much social unrest is re-vindicative rather than revolutionary, since the discontent of long-suffering groups often is galvanized into action only when customary entitlements are threatened or denied."11 Just like extended individual rationality, extended social rationality consists of social nonmoral good and the instrumental rationality to achieve it. Social rationality is the definition which Railton gives for moral rightness. That is morally right which is socially rational i.e. conducive to social nonmoral good. What is conducive to social nonmoral good is determined by naturalistic properties and facts. Thus, moral rightness as social rationality is a naturalized criterion, in the same way individual rationality is a naturalized criterion.
Expectations
We must now see whether this naturalized social criterion can describe and predict social-moral phenomena as well as individual rationality can for individual phenomena. If the social rationality theory of moral rightness is true, then we expect to observe the following. Just as individuals who do not achieve their objective interests get dissatisfied, so does a society which does not achieve its objective social interests get dissatisfied, which we assume manifests as social unrest. Social rationality takes into account every individual's nonmoral good. So the more social rationality is approximated, the more all individuals' interests are satisfied, the less motivated they are to instigate reforms and revolutions. So we expect to observe that the more a society approximates social rationality, the less social unrest there will be. If a given society is very stable for a long time, we can take that as evidence for that society being very socially rational (though it is not necessarily conclusive evidence, as we shall see). The independence of individual nonmoral good from subjective beliefs explained many phenomena. For example, a person can be dissatisfied, despite believing that what they are currently doing is good for them, because in objective fact it is not good for them. We expect that certain social phenomena can be explained in the same way. That if in objective fact a society does not approximate social rationality, then there would be social unrest regardless of prevailing beliefs. This expectation is confirmed by examples of authoritarian regimes with overwhelming propaganda. The regime may succeed in brainwashing people. But people's beliefs would not prevent them from actually suffering objective problems like hunger, lack of security, arbitrary arrests and executions. Along this line, we saw that individuals could be unconsciously guided towards their objective interests by feedback mechanisms. That is, individuals can tend to gravitate towards their objective interests, without knowing of them or perhaps even holding beliefs contrary to them. In the same way, objective social good i.e. social rationality can direct social behaviors and decisions towards their better approximation, regardless of the beliefs commonly held in that society. So social improvement can sometimes be explained independently of societal beliefs.
Social feedback mechanisms would predict in areas where social feedback mechanisms work better, norms would better approximate social rationality and therefore also be better established and stable. And vice versa, in areas where social feedback is not effective, norms would not approximate social rationality well and would be unstable. Feedback mechanisms would work well in areas where people share similar interests or interests that do not conflict, because there is a solution (a system of norms) compatible with satisfying most or everyone's interests. Furthermore, what counts as positive or negative, a gain or a loss, is clear i.e. clear feedback. Good feedback also requires there be no great power imbalance, so that pressure can easily and strongly be exerted on behalf of most or all parties' interests. This yields impartiality. Implementation of the desired response is greatly facilitated where individual compliance greatly encourages general compliance, and where behavioral adjustment is straightforward and easy. To increase the probability that the proposed response will be introduced in the first place, the suggested norms must be advantageous at all scales, from small groups to large (so that they are first introduced in small groups, and then eventually are institutionalized into larger groups). According to Railton, history bears out this pattern, that the most well-established norms are in areas satisfying the above criteria. For example, norms forbidding aggression, theft and the breaking of promises have been staples of societies, and seem to approximate social rationality well. Conversely, there are not many well-established norms are in areas satisfying the negations of these criteria. For example, norms forbidding coercion by powerful parties have historically not been well enforced, precisely because of great power imbalances. Examples include slavery, class or sex inequalities, authoritarian regimes…
We expect that societies will tend to become more socially rational over time, because parties suffering social irrationality will be motivated to fight for their interests, and will do so if they are powerful enough. Pressure from various parties' active efforts at defending their own interests is one of the primary social feedback mechanisms. Railton calls this the Generality trend, that norms become increasingly undiscriminating, equal and inclusive of marginalized groups. History bears this out, with “mobilization of excluded groups to promote greater representation of their interests include the rebellions against the system of feudal estates, and more recent social movements against restrictions on religious practices, on suffrage and other civil rights, and on collective bargaining.”12 Railton also invokes the historical tendency of humans to associate in increasingly large groups. We began by living in smaller groups the size of families or tribes. Then we formed villages, cities and eventually nation states. The increase in group size itself is not a result of impartiality, merely a conquering by the stronger of the weaker, who are then incorporated into expanded territories. But in these increasingly large territories, the conquered eventually fight for their interests to gain equal status with the conquerors, and this is evidence confirming impartiality. This expected moral progress together with Social Rationality’s objective social good would seem to imply that societies would progress towards the same ideal i.e. moral convergence should increase over time.
Evaluation
But we do not observe moral progress and increasing moral convergence. There is some overlap, but moralities in different cultures and religions still have significant differences. And there have been instances where societies seemed to regress away from social rationality. To this Railton responds: Just because something is true, doesn't guarantee that the truth will out. Scientific realism does not imply scientific progress or convergence. Scientific truths do not guarantee that all societies will accept them or succeed in approximating them. But this does not remove or change the fact that natural phenomena reliably behave a certain way. In the same way, moral realism implies neither moral progress nor convergence. There can be moral truths, which societies can accept and try to accord with or not. But this doesn’t change the fact that certain stimuli reliably produce certain responses in human individuals and groups. Furthermore, just because a pattern exists doesn’t mean it must dominate. For example, social rationality would not win out in cases where there are great power imbalances between parties to a conflict of interest. Those stronger would use any means to keep the weaker down, thus compromising social rationality. Extending this, it is naive to think that society must progress towards perfect social rationality. It is more plausible to expect that it will move towards an equilibrium between parties powerful enough to defend their own interests. The weak will get no share.
I have already evaluated Railton’s theory incrementally at each of the concepts leading up to Social Rationality. So I will just briefly evaluate the theory as a whole. We have seen at each step that the theories and definitions seem to fit well with the moral data. Another advantage is that although social rationality is highly explanatory, it is definitely not a theory that merely endorses the status quo. Its analyses can demand reform. If a norm cannot plausibly be understood as a mechanism conducive to social rationality, then it should be questioned. If a norm was created, reinforced, taught under conditions unconducive to social rationality, then it should be carefully examined. For example if a certain moral norm was created and enforced under conditions of large power imbalances, we might reasonably suspect that the resultant norms were merely enforced by coercion and not socially rational. Of course, metaethical naturalism presupposes ontological and epistemological naturalism. And while its coherence is a virtue, metaethical naturalism stands and falls with ontological-epistemological naturalism, the advantages and disadvantages of which are beyond the scope of this paper. Another advantage is that moral (social) normative facts are based in individuals’ normative facts, which are in-principle motivating. This establishes a connection between normativity and motivation which fits with our intuitions about how compelling moral concerns should be. But some may be repelled by the normative force of moral reasons being ultimately based on individuals’ objective interests. They may find such a view too selfish and brutish e.g. Kant would prefer the universal respect for rational autonomous agents as motivational force. Railton cannot allay such feelings since, for him, morality is not intrinsically valuable. Morality gains its authority from the fact that humans have biological needs that require a strategy for satisfaction, a strategy which must account for the fact that they live in groups and must live together in a way that satisfies everyone sufficiently, at risk of social instability otherwise.
References
Fisher, Andrew. Metaethics: An Introduction. Taylor and Francis, 2014. https://doi.org/10.4324/9781315711393 .
Potochnik, Angela, et al. ‘Experiments and Studies’. Recipes for Science : an Introduction to Scientific Methods and Reasoning. Routledge, Taylor & Francis Group, 2019.
Railton, Peter. ‘Moral Realism’. The Philosophical Review, vol. 95, no. 2, 1986, pp. 163–207. JSTOR, https://doi.org/10.2307/2185589. Accessed 20 Feb. 2023.
Sayre-McCord, Geoff. ‘Metaethics’. The Stanford Encylopedia of Philosophy, edited by Zalta, E. N., Stanford University, 2014 summer edition. https://plato.stanford.edu/cgi-bin/encyclopedia/archinfo.cgi?entry=metaethics .