1 Introduction

When philosophers talk about “moral motivation,” the basic phenomenon they seek to understand is that people are generally inclined to do what they believe it right to do. The predominant philosophical approach to this topic has been to focus upon what people believe in order to explain what they do, for example, by appealing to the nature of the properties that figure in moral beliefs or by exploring the connection between motivation and moral judgment. Little attention has been paid, however, to the issue of how we can be motivated to act consistently with our beliefs. How to understand the quality or strength of moral motivation?

Notwithstanding the complexity of the mainstream philosophical debate on motivation, I argue that the above approaches do not fully capture what is at stake when we give our allegiance to an ethico-political view or defend the values of a given tradition. This is because over and above what people believe and what they do, there remains the issue of what people aspire to be. Over and above the gap between what we believe and what we do, there is a gap between how we are and how we wish to be that is crucial to the phenomenon of moral motivation. To clarify this point, this paper argues that a proper account of moral motivation needs to cross the boundaries of mainstream approaches to allow for the distinctively emancipatory dimension of being motivated that is often overlooked in current debates.

In so arguing, I focus on what I will call the “thick” moral phenomenology of a philosopher whose views have been largely neglected in metaethics: Charles Taylor.Footnote 1 Already in Sources of the Self, Taylor criticizes the then popular ethical theories for having a too narrow focus on “what it is right to do rather than on what it is good to be” (1989, 3). In the same spirit, he endorses Iris Murdoch’s critique of the “narrowness” of analytic moral philosophy (Taylor 2011, 3–23). This paper revitalizes this discussion by reflecting on the phenomenology of motivation. What does the experience of motivation consist in? What is it to be morally motivated?

The main points are (1) that moral motivation has a distinctive emancipatory dimension which has been largely neglected in mainstream debates; (2) that the issue of emancipation can only be adequately conceptualized at the intersection of normative ethics and metaethics; (3) that a full-blooded account of motivation must incorporate a phenomenology of motivational experience, which in turn requires (4) extending the concept of motivation beyond a narrow definition to include such notions as meaning, articulation, identity, and freedom, and (5) criticizing conceptions of motivation that are blind to or take for granted the quality of motivation; and thus (6) providing the resources for a thick conception of motivation that breaks new ground by overcoming the existing boundaries between normative ethics and metaethics.

As I am using the term, ethical emancipation—the act or process of being liberated through morality—is something stronger than what is generally described as “moral motivation,” for the reason just given: the latter typically indicates a sense of what it is good to do, whereas emancipation also covers what it is good to be. “Emancipation” is often defined in negative and passive terms, as a process of being set free from another person’s control or from legal, social, or political restrictions, for example, in phrases such as the “emancipation from slavery” or the “emancipation of women.” What I have in mind here is slightly different: ethical emancipation is a distinctively positive and active process of reaching towards moral growth, maturity, self-understanding, and self-realization. It involves having a sense of a higher motivation, as opposed to a more depraved one, which we aim to grow beyond in pursuit of the higher standpoint.Footnote 2

By focusing on motivation and emancipation as correlative concepts, I am explicitly reaching beyond the specialist framework of metaethics—the sub-discipline in which issues of motivation are generally debated—to make room instead for a moral-phenomenological analysis of the different ways in which people can feel motivated to do what they judge right. Including the notion of emancipation in the reflection on motivation calls for a wider approach than the traditional metaethical analysis of the nature of moral judgments, to the extent that this notion is characteristically expressed through moral experiences and sensibilities whose value and resonance cannot be exhausted by a (non-)cognitive analysis of moral beliefs. This, of course, is not to say that broad metaethical reflections do not impose constraints on what motivation consists in, or that such constraints are unimportant.Footnote 3 The point is rather that there is more to the philosophy of motivation than the philosophy of beliefs and truths, and that there are further questions to be raised about the nature of motivation by adopting a broader perspective.

Thus taking a distance from metaethical discussions of motivation, I start from a highly controversial discussion in Sources of the Self, in which Taylor poses the question as to whether we, late-modern Westerners, “are not living beyond our moral means in continuing allegiance to our standards of justice and benevolence,” expressing his concern that an increasingly secularizing culture might be aiming higher than its moral sources can sustain (1989, 517). I first elaborate on this question and show why Taylor’s concern is somewhat of an uneasy fit with the broad philosophical debate on motivation (§ 1). I continue to explain his unusual voice in the debate by drawing a contrast between Taylor’s key concept of moral sources and the more familiar notion of moral facts (§ 2). Against the background of this contrast, I turn to Taylor’s view on the nature of human language to explain how it informs his accounts of motivation (§ 3), emancipation and realism (§ 4), and freedom (§ 5). I finally return to Taylor’s doubt about secular sources and draw some conclusions for the philosophy of motivation as a whole (§ 6).

2 Taylor’s Question

Perhaps the most vital concept of Sources of the Self (as the title indicates) is that of a “moral source.” The central feature of moral sources in Taylor’s sense is that they provide motivational strength: “Moral sources empower. To come closer to them, to have a clearer view of them, to come to grasp what they involve, is for those who recognize them to be moved to love or respect them, and through this love/respect to be better enabled to live up to them” (Taylor 1989, 96). The understanding of motivation in terms of empowering sources provides the context for Taylor’s diagnostic analysis towards the end of the book, in which he distributes the moral sources of the modern self into three large domains: one theistic, a second one that centers on “a naturalism of disengaged reason” and a third area which “finds its sources in Romantic expressivism or in one of the modernist successor visions” (1989, 495). These three general types of moral source set the stage for Taylor’s highly controversial claim that the sources of disengaged reason and Romantic expressivism are in a quite different situation from theistic ones.Footnote 4 As he says:

Theism is, of course, contested as to its truth. Opponents may judge it harshly and think that it would be degrading and unfortunate for humans if it were true. But no one doubts that those who embrace it will find a fully adequate moral source in it. The other two sources suffer a contestation on this score. The question is whether, even granted we fully recognize the dignity of disengaged reason, or the goodness of nature, this is in fact enough to justify the importance we put on it, the moral store we set by it, the ideals we erect on it. (1989, 317)

How to make sense of this concern? Ultimately, Taylor’s position is that “high standards need strong sources” (1989, 516). In the lecture “A Catholic Modernity?” he explains this view by adding that we are living in an “extraordinary moral culture,” historically speaking, in which “suffering and death, through famine, flood, earthquake, pestilence, or war, can awaken worldwide movements of sympathy and practical solidarity” (2011, 177). Reflecting on the Christian roots of this culture, he notes that the practice of giving aid to those in need gradually became part of a much larger movement, where it comes to be motivated not merely by Christianity but by a great variety of moral outlooks. Given this history, Taylor concludes, “our age makes higher demands for solidarity and benevolence on people today than ever before” (2011, 182).

Against this background, his concern is that an increasingly secularizing culture might be aiming higher than its moral sources can sustain. To be sure, we live up to the standards of solidarity and benevolence to the extent that we do because they have become part of what we understand as a normal, civilized human life. However, Taylor tells us, this makes a rather fragile motivation: “A solidarity ultimately driven by the giver’s own sense of moral superiority is a whimsical and fickle thing. We are far, in fact, from the universality and unconditionality which our moral outlook prescribes” (2011, 182). In other words, if merely having “appropriate” beliefs is no guarantee for sustaining affirmation of the high demands of solidarity and benevolence, how, then, Taylor asks, will we face the moral challenges of our age? This is now the context in which he expresses his concern that secular sources might be less adequate than theistic ones for the affirmation of human beings as worthy of solidarity and benevolence. And yet, it emerges from the above discussion that Taylor’s doubt about secular motivation is a question rather than a statement. How to make sense of this question in terms of the philosophical debate on motivation?

The main problem in interpreting Taylor’s question is that it encompasses elements of two categories in analytic moral philosophy which are typically seen as separate: normative ethics and metaethics. That is, for many philosophers, it is simply good practice to distinguish between “first-order” or normative questions within morality and metaethical or “second-order” questions about morality. The claim that murder is wrong is a normative claim; the claim that this belief cannot be true is a metaethical claim. While this distinction has certain pedagogical merits, it seems rather unhelpful or even distorting when debating questions about motivation, in which normative issues (How should I live?) are inextricably intertwined with metaethical ones (Are my moral views true?). As Sayre-McCord puts this point, “despite the abstract and deeply controversial nature of metaethics, its central concerns arise naturally—perhaps even inevitably—as one reflects critically on one’s own moral convictions” (2014). The problem with Taylor’s question, then, is that it is neither strictly normative nor strictly metaethical, as it at once makes a claim about the “normative” issue what solidarity and benevolence require and about the “metaethical” issue what ontology makes best sense of our moral commitments. For present purposes, his unusually broad perspective can be seen as preparing the ground for the idea that motivation involves more than what we ought to believe and do. It involves a broader sense that includes what it is good to be.

With these points in place, it is hardly surprising that it is precisely this broader understanding about which Taylor expresses his concern that is taken for granted in mainstream debates on motivation. As Rosati explains, contemporary ethical theories generally “take no position regarding the exact strength of moral motivation,” while also “taking it for granted that moral judgments dependably, if not unfailingly, motivate, that they effectively influence and guide how people feel and act” (2016). These observations—that most ethical theories remain silent about motivational strength and tend to see moral motivation as a consistent and reliable phenomenon—point to a peculiar gap within the literature, namely, that surprisingly little has been said about the phenomenal character of motivation; about the different ways in which people can feel motivated to do what they judge right. Rosati discusses a wide variety of views and discussions concerning motivation,Footnote 5 but, she continues, for all the metaethical credentials these theories are able to produce, “the motivation all or some people feel to do what they judge right might be extraordinarily weak” within these theories, leaving it an open question whether moral motivation is indeed as stable as is commonly supposed. This is now the uncertainty in which Taylor’s concern for adequate sources takes root.

The discussion so far puts us in a predicament: if Rosati is right to argue that much contemporary philosophy takes it for granted that moral judgments motivate, then there seems little conceptual space left within this framework to make sense of Taylor’s question, for he precisely rejects this assumption by allowing the possibility that we lack the motivational strength to live up to the moral demands we impose on ourselves. It would seem, therefore, that we need a quite different approach to motivation if we are to examine Taylor’s concern more closely, as we need to resist both the consensus that normative concerns must be separated from metaethical ones and the habit of investigating motivation in terms of the beliefs we hold. What we need, in other words, is to shift the focus away from moral epistemology onto moral phenomenology.

3 Moral Sources are not Moral Facts

Because Taylor’s view does not fit well with the many and various philosophical taxonomies and classifications of theories, it might be helpful to contrast his view with what it is not.Footnote 6 Taylor’s key concept of a moral source is emphatically not what philosophers call a “moral fact.” Somewhat confusingly however, Taylor’s moral phenomenology also has some affinity with the metaethical theory that champions moral facts, called “robust realism.”Footnote 7 It seems fair to say that this position is (at least partly) an exercise in moral phenomenology, because, like Taylor, robust realists aim to be sincere and true to the character of our moral experience (cf. FitzPatrick 2008; Enoch 2011). Moreover, like robust realists, Taylor has been fighting naturalism in ethics from the outset of his career and he has admitted to be arguing for “a kind of moral realism” (Taylor 1991, 242; Meijer and Taylor 2020, 990). To clarify his unusual position, I want to briefly zoom in on one specific aspect of robust realism: its self-declared inarticulacy about the source of value. As will emerge, this feature stands in direct opposition to Taylor’s understanding of moral sources.

For moral-phenomenological purposes, it is important to note that the robust realist does not engage directly with moral experience but turns instead to the judgments we make on account of them. That is, robust realism argues that moral experience is best understood in terms of moral judgments, which, in turn, are best understood as factual statements. In this respect, moral judgments and ordinary beliefs about the world are seen as fundamentally alike as they both seek to describe facts which reflect truths not of our own making. As FitzPatrick explains this view: “Ethical claims purport to state facts […] and so are straightforwardly true or false in the way that other purportedly fact-stating claims are, by accurately representing the facts or not” (2008, 161).

For the robust realist, moral facts are “brute facts about value that are not further explicable” (FitzPatrick 2008, 194). In this respect, Shafer-Landau makes it clear that being a moral realist just is to be committed to “a set of brute facts for which no further explanation is available,” as a basic metaphysical reality: values just are “a brute fact about the way the world works” and there may not be much more for philosophy to say here (Shafer-Landau and Cuneo 2007, 158; Shafer-Landau 2003, 48). The implication here is that robust realism sees values either as purely abstract objects, perhaps like mathematical entities such as prime numbers (Enoch 2011, 203) or as properties that “inhabit” the world as entities that are there anyway—perhaps on a par with geological or biological properties (Shafer-Landau 2003, 63). Shafer-Landau in particular rejects any demand for an explanation of the source of value: correct moral judgments are just true and there is nothing that makes them true (2003, 47–48, 96–97).

From Taylor’s perspective, this kind of inarticulateness is deeply confused. In fact, it is rooted in a form of self-delusion which completely ignores what Taylor has been calling the “constitutive power of language.” This is the central topic of his book The Language Animal (2016), which adds an extra dimension to his moral phenomenology by invoking the relation between language and meaning. Before embarking on that topic, however, we need first ask: What is the nature of moral sources and how are they different from moral facts?

As we have seen, Taylor’s account of moral sources specifically invokes the question of how we are motivated by (our recognition of) value. He describes the value commitments implicit in people’s motivation both as moral sources and as “constitutive goods.” This brings us to Taylor’s distinction between two types of moral goods. He uses the term “life goods” to articulate “what actions, modes of being, virtues really define a good life for us” and employs the notions of “constitutive goods” and “moral sources” to depict the “higher-order” goods that inspire life goods (1989, 92–93; 1997, 173). By classifying these two types, Taylor seeks to capture the sense that moral agents qua moral agents not only recognize multiple goods but feel that they have to rank them. The idea of moral sources as higher-order constitutive goods expresses Taylor’s view of ethics as involving “a range of ‘values’ that are essentially understood to be on a different level, to be in some way special, higher, or incommensurable with our other goals and desires” (2003, 308). That is, we acknowledge “second-order qualitative distinctions which define higher goods, on the basis of which we discriminate among other goods” (1989, 63).

Robust realism’s insistence on brute moral facts flies in the face of Taylor’s point that espousing a life good without reference to constitutive goods or sources would be dealing with a “compulsion,” not a moral obligation (2011, 9). To make this point, he draws a parallel with the neurotic necessity to “wash one’s hands” or to “remove stones from the road” (2011, 9). That is, just as the acts of washing one’s hands and removing stones from the road become unintelligible when separated from an underlying motivation that includes their point (for example, to have clean hands and to keep the road passable), so too does a life good require an explanation of why one should live up to the norms that it brings about. This further clarification, then, articulates what Taylor calls “moral sources.” As he argues, “a moral obligation comes across as moral because it is part of a broader sense which includes the goodness, perhaps the nobility or admirability, of being someone who lives up to it” (2011, 9).

Regardless of the content of one’s moral orientation, the sense is that (lower-order) life goods need crucial reference to (higher-order) constitutive goods or sources by virtue of which these life goods are recognized as “good.” For example, if I see devotion to God as a crucial part of the good life, I am articulating a life good. But, Taylor argues, this makes sense only in the broader context of seeing ourselves as creatures of a loving God. Similarly, universal human rights cannot be separated from a worldview in which human beings are of equal worth. For the same reason, it would not be intelligible to espouse life goods such as a good marriage, a happy family, and a successful career without some underlying recognition of work and family as the main sources of a full human life. These examples indicate that life goods define what we find of moral importance, whereas constitutive goods explain why we believe this to be so. But—and this point is essential—this requires words. Moral sources as constitutive goods must therefore be seen not just as crucial motivations to act in some ways rather than others, but also as deeper articulations of one’s sense of value.

Herein lies the crucial contrast between Taylor’s articulation view and “neurotic” robust realism: the role attributed to language for ethical theorizing. The robust realist understands moral language in terms of the recognition of moral facts. These facts are independent in the sense of being self-standing, and the role of language is to designate and describe them. By contrast, Taylor’s articulation model stresses the importance of linguistic expression not to describe some external reality, but to show that the moral sources implicit in our motivation need to be articulated to appear at all. From this perspective, making sense of motivation requires more than reflecting on the fact-stating logic of moral judgments; it also involves reflecting on the moral meanings to which our judgments respond. As will be shown, this calls for a further distinction, between the different conceptions of language in the background of the vocabulary of moral sources and moral facts. At this stage of the argument, however, it suffices to note that moral sources are distinctly different from moral facts precisely because they are part of the realm of experienced meaning. Because of this, their existence “doesn’t precede the articulation but comes about through and with it” (Taylor 2016, 188).

4 Articulation I: Motivation

Taylor’s view on language has come a long way. Already in the introduction to Philosophical Papers, he makes it clear that his thought at that time is too premature to grasp the connection between language and meaning in full (1985a, 11–12). The introduction to Philosophical Arguments, ten years later, repeats this uncertainty. In this text, Taylor takes his cue from the Romantic “constitutive” theory of language, which sees language “not primarily as an instrument” (say, to describe moral facts) but rather as “what allows us to have the world that we have” or as something that “makes possible the disclosure of the human world” (1995, ix). On the one hand, it seems clear that this is the context in which Taylor’s plea for articulation starts to make sense. On the other hand, when he admits that “there is a combination here of creation and discovery which is not easy to define” (1995, ix-x), it also seems we are still not making much progress.

The publication of The Language Animal in 2016 (two decades later) can be seen as Taylor’s most matured position on this topic. In this book, he seeks to explain the relation between language and meaning by separating what he calls “life meanings,” which we share with other animals, from “human meanings” or “metabiological meanings,” that is, meanings on a moral or spiritual level, having to do with what are seen as the highest goals or the best way of life (2016, 91). One crucial feature of such meanings is that they invoke our self-understanding:

If what I seek is a meaningful life, or a profound sense of peace […] what I’m after can’t be captured in some objectively identifiable pattern. In order to see what’s at stake here, one has to get inside the language of self-description, catch on to what a meaningful life is for me […] what I mean by a sense of peace. (Taylor 2016, 91–92, my emphasis)

Taylor’s claim that the articulation of human meaning cannot be separated from the way in which we understand ourselves is rooted in his phenomenology of motivational experience, as elaborated in the early paper “Self-Interpreting Animals.” In reflecting on our “experienced motivation,” that is, “our feelings, emotions, desires,” Taylor distinguishes between three types of emotion that move us to act (1985a, 47).Footnote 8 Together these three types set the stage for the central claim that “our interpretation of ourselves and our experience is constitutive of what we are” (1985a, 47). The first type of emotion is what he calls “immediate feelings,” such as physical pain (1985a, 72). Pain is immediate in the sense that it does not require language to exist, which is why we can easily imagine non-human animals having this experience. A second type of motivation involves “import emotions,” an experience which is different from the brute reaction in immediate feelings because it involves “a sense of our situation” (1985a, 48). Taylor’s example of an import emotion is fear or shame, emotions through which we experience our situation as bearing a certain “import”—its being frightening or shameful. This means that import emotions are more complex than immediate feelings such as tickles, itches, and pain: to be afraid or ashamed is not just to experience a certain feeling but to be aware of the situation in a particular way. Whereas a situation is painful “just because we feel pain in it,” the feeling of fear/shame is a response to what has happened, which makes explicit some judgment about the situation that gives the emotion its character (1985a, 48–49). Thus Taylor explains that import emotions are best understood as “affective modes of awareness” (1985a, 48). The important point is that such modes of awareness could not exist for us without the affect: import emotions “cannot be neutral, cannot be something to which we are indifferent, or else we would not be moved” (1985a, 48).

Because import emotions can either remain implicit or be expressed in explicit language, they apply to human and non-human animals alike. Yet there is a third type of emotion that, according to Taylor, emerges from our nature as uniquely linguistic animals: strongly evaluative motivations. What distinguishes this class of emotions is that they are what Taylor calls “inherently reflexive:” they involve not just explicating the imports concerned, but also referring to “the life of the subject” (1985a, 67). This is where the connection between articulation and self-understanding starts to emerge. With his key concept of “strong evaluation,”Footnote 9 Taylor seeks to capture the sense that human beings understand themselves not by simply having certain desires but by evaluating their desires in terms of their worth. In this way, he employs the concept of strong evaluation to highlight that human beings experience some of their motivations as inherently more worthy than others:

It is because this involves ranking motivations that I speak of it as strong evaluation. It means that we are not taking our de facto desires as the ultimate in justification, but are going beyond that to their worth. We are evaluating not just objects in the light of our desires, but also the desires themselves. (Taylor 1985a, 66)

Shame and pride are strongly evaluative emotions because they need crucial reference to one’s self-understanding. For example, I might be proud of my success as a bank robber, while you are really ashamed of me. Shame and pride refer to activities and ends which are worthy or unworthy. As import emotions, they involve a response to and make explicit some judgment about the situation or some state of affairs. But as strong evaluations, they further reflect how this response/judgment relates to one’s self-understanding in terms of worthiness. Strong evaluation thus involves attributing to different motivations their place in the life of the subject:

Implicit in this strong evaluation is thus a placing of our different motivations relative to each other, the drawing, as it were, of a moral map of ourselves; we contrast a higher motivation with a baser, more self-enclosed and troubled one, which we can see ourselves as potentially growing beyond, if and when we can come to experience things from the higher standpoint. (Taylor 1985a, 67)

Blakely makes the important point that according to Taylor, even our immediate feelings and import emotions tend to be colored by the interpretations that comprise the moral map that defines our identity: “an emotion like fear or even pain might appear to us as cowardly or heroic, exulting or masochistic” (Blakely 2013, 394). The upshot is that “human life is never without interpreted feeling; the interpretation is constitutive of the feeling” (Taylor 1985a, 63). In this sense, then, the strong evaluations by which we interpret ourselves are constitutive of what we are as persons.

5 Articulation II: Emancipation

Strongly evaluative motivations are expressive of the self-interpretations of human identity. They judge not merely some feature of reality, but also the significance of one’s own desires, actions, and identity as part of that reality. But what are the broader philosophical implications of this distinctive type of articulation? What is actually happening when we find the right words for our feelings and motives? As the later Taylor makes plain, our nature as “language animals” brings out something essential about human agency indeed:

The “right word” here discloses, brings the phenomenon properly into view for the first time. Discovery and invention are two sides of the same coin; we devise an expression which allows what we are striving to encompass to appear. This is a crucial facet of our language capability, which I will call “articulation”. (2016, 178)

As noted above (in § 2), the moral sources that our strong evaluations seek to define can only exist or appear for us through linguistic or other forms of expression (such as painting, sculpture, music, dance, ritual, etc.). More needs to be said, however, on the nature of specifically verbal expressions. Unlike the meanings involved in expressing our immediate feelings, strong evaluations are such that articulating something with the term “shameful” inflects our sense of meaning in a new direction. Because of this, new expressions “open up new ways of being in the world” (Taylor 2016, 189). We are concerned here with a particular constitutive force of strong evaluations, which puts us squarely in the domain of motivation because it reflects on us as language animals seeking moral orientation through articulation:

Prior to the articulation, the as yet unnamed import may be felt in a diffuse, unfocused way, a pressure that we can’t yet respond to. After articulation, it becomes part of the explicit shape of meaning for us. As a result it is felt differently; our experience is changed; it has a more direct bearing on our lives. […] Articulation here alters the shape of what matters to us. It changes us. (Taylor 2016, 189, my emphasis)

The upshot of this view is that the articulation of meaning creates not merely new realities, but also a new sense of self as part of that reality. When I come to understand my agitated state as envy or jealousy, I am already living it differently. I have taken the first step out of confusion; my situation already has a shape for me. In this way, new articulations allow the world to impinge on us, to move us, in new ways. That is why Taylor calls them “constitutive.” But again, this particular kind of articulation requires words. An initially vague “unpleasant” feeling emerges as a recognizably distinct experience of envy, jealousy, vanity, indignation, remorse, or whatever, when we find the right words. The strongly evaluative descriptions carry the constitutive force.

All of this evaporates in an account of brute moral facts. Describing one’s motivation in terms of facts which are anyway there (on a par with mathematical, biological or geological properties) may bring clarity to the world but it does not alter it. By contrast, when I see that the issue which really concerns me is envy or jealousy, I am clear about something which is not independent of my clarified vision. What I have clarified is my sense of what matters to me, which changes the shape of what matters to me. Perhaps I sense that I have been feeling this all along, and just now am recognizing it, but this recognition gives a new force and clarity to this meaning. As Taylor puts this point, the discovery has “motivational force” (2016, 191). It is this kind of articulacy, which changes its object, that is crucial to moral language as Taylor understands it. Evidently, the experience of discovery alluded to here is emphatically not like discovering a mathematical theorem, a new animal species or how heat flows from the earth.

We can make sense of this distinction, between describing independent facts and introducing new constitutive expressions that alter the field of meanings, by bringing out the different semantic logics involved. The first is the robust realist’ one: we have a strong moral intuition, say, that one should not lie, and we call this a moral fact. This follows the logic of a correspondence theory of truth: find an account which corresponds to an independent reality. This is what Taylor calls a “designative” logic (2016, 191). The second one is the “constitutive” logic where introducing a new term “reorders or reshapes the field of phenomena it helps to describe” (2016, 192). In short, in the designative logic the phenomenon comes first, then the term to describe it, whereas in the constitutive logic the new term and the reality it describes are concurrent.

And yet, when Taylor states that strong evaluations involve ends or goods which “stand independent of our own desires, inclinations, or choices” (1989, 20) it seems clear that these experienced feelings refer beyond themselves to orders of reality which we could be getting wrong. This is Taylor’s view indeed: “our perception of value makes claims about the way things are in some sense independent of us, claims which could turn out wrong” (2016, 193). While this point appears to commit him to the designative logic of robust realism, a crucial difference remains. The disanalogy is that the epistemic gain in the realm of meanings comes through “the change in felt intuitions which the words bring about in us” (Taylor 2016, 193, my emphasis), not through a changed understanding of “the way the world works” in a quasi-scientific sense, which can be dispassionately contemplated (Shafer-Landau 2003, 48, 63).Footnote 10 The point is that when it comes to our strongly evaluative judgments, we could always be deceiving ourselves—for instance, when I pride myself for acting out of generosity while I am really acting out of spite.

Here Taylor’s highly original view of the distinctive kind of facticity, objectivity, and independence involved in moral judgments comes to the fore. He rejects the view that facts and values lie in different realms as question-begging, precisely because this requires the additional assumption that moral statements are not really “factual” (Taylor 2016, 193n14). Rather, his view is that the independence involved in our strong evaluations is utterly different from the self-standing objectivity of moral and natural facts alike. As we have seen, what Taylor calls human meanings are strongly dependent on our modes of articulating them. But we also saw that this does not mean that issues of truth cannot arise in relation to these meanings: whereas we cannot feel pain mistakenly, we can be wrong about our strongly evaluative motivations. To return to the earlier example, where I am proud of my success as a bank robber, your being ashamed of me may begin to work on me to the point that I come to see that this is not a worthy achievement at all. The point being: there is a reality here, to which our judgments are answerable. Strong evaluation thus involves goods which stand “independent” of our own desires, in that our experience of these goods can be faulty and need correction.

In this sense, then, there is supposed to be a truth of the matter underlying strong evaluations. In this sense, that is, “our moral reactions suppose that they are responses to some reality, and can be criticized for misapprehension of this reality” (Taylor 2011, 297). In this sense, in other words, human beings are receptive to the demands of reality itself. This is now the shape of what might be called Taylor’s “moral-phenomenological realism.” This realism is informed, first, by a demand for “undistorted, not-too-self-indulgent self-description,” and second, by a claim to “strong value, that is, value independent of our recognition,” which together put a burden on us to “get things right” (Taylor 2016, 195).

We already observed the categorical difference between a moral source and a moral fact: the former is phenomenologically richer and thicker as it involves not just truth but also meaning. For the same reason, it is also true that strong evaluations are not just moral beliefs. As they involve affective imports of significance, getting it right in strong evaluation is more than a matter of having true moral beliefs by getting the facts right. Having true beliefs is simply not enough. Grasping our strong evaluations also requires being able to grasp how such beliefs relate to our sense of our life as a whole and the direction it is taking as we lead it.Footnote 11 From this perspective, robust realism can be criticized not for insisting on the representational nature of moral judgment, but for upholding a picture of moral agents as detached observers, which seems to leave little room for the identity-constituting force of moral judgment. While there are cases where the robust realist’ phenomenology of moral judgment seems right (for example, when we are assessing other people’s conduct) the identity of moral agents understood as “epistemic judges” or “quasi-scientists” does not appear to be playing a significant role to the extent that their knowledge is “not constitutive of moral facts” in reflecting “a truth not of their own making” (Shafer-Landau 2003, 17; Enoch 2018, 215).

This means that, from Taylor’s perspective, there are at least two dimensions in which our moral language is answerable to the world. The first, “factual” dimension concerns “the adequacy of our descriptions, of self and others, and our situations” (2016, 194). However, Taylor’s realism does not stop here—at the metaethical question of whether our moral claims are true—as it involves a more normative concern for who we are and aspire to be. This is why he adds a second, “normative” dimension, which concerns “the validity of what we value, of the norms, goods, virtues we want to seek” (2016, 194–195). For this reason, Taylor’s moral realism cannot stop at simply listing a series of brute moral facts, such as “that we should care about our future well-being” or “that we should not humiliate other people,” and simply declare that such facts are not further explicable (Enoch 2011, 1). From within Taylor’s view, it is a kind of discussion-stopper to insist on brute facticity in this area. Without a further articulation of the wider normative significance of values such as dignity, respect, solidarity, and benevolence, which make sense of the above judgments, these facts do not explain all that much.

In conclusion: yes, moral judgments are factual statements but

  1. (1)

    These have to be understood from within a constitutive logic, not a designative one, which means…

  2. (2)

    …that they have a unique kind of independence and objectivity in constituting human identity, which in turn implies…

  3. (3)

    …that they involve factual questions of adequacy as well as normative issues of validity.

Taken together, these points show that getting it right in moral matters is an emancipatory matter more than anything else; the drawing, as Taylor puts it, of a “moral map” of both the world and our place in it. With these points established, the question that remains is: How do we proceed in such moral mapping?

6 Articulation III: Freedom

What are the upshots of Taylor’s phenomenology for understanding moral motivation? Perhaps the most fundamental point of Taylor’s view is that the constitutive force of our strong evaluations has transformational power as well. As we have seen (in § 2), one way of making sense of the “landscape” or “moral space” (1989, 25–52) which our strong evaluations seek to define includes the explicit articulation of constitutive goods. Such higher-order goods appeal to some sense of (non-)human life as worthy of our concern, which might be characterized in terms of an account of human dignity, evolution, God, deep ecology, and so on. Taylor calls such accounts “moral ontologies” (1989, 8–10) or “etiological” accounts (2016, 196) because they offer fundamental explanations for our experienced motivation; they each present an explanatory background which supposedly underlies our strong evaluations. As noted, the articulation of moral sources as constitutive goods by no means offers a merely dispassionate explanation: as moral ontologies, they constitute and color the meanings in a way that gives them a profoundly normative character. Consequently, any change of description effects a change in the reality, that is, “the pattern of meanings we live by, the ‘landscape’ as we live it and feel it” (Taylor 2016, 197). In Taylor’s view, then, articulating the sources of motivation implies a double call on us: the factual one “to get it more clearly in focus” and the normative one “to live up to this sense of what is important” (2016, 196). These two transformations involve changing ourselves in these two dimensions in a deeply holistic way: part of the fruit of articulating better is getting better, and vice versa.

As Taylor envisages it, the debate on moral motivation is thus ultimately a debate on possible or impossible ethical transformations. It is a debate on what ethical emancipation demands. This takes us to the heart of the issue of moral sources: How do these sources strengthen us? How to bring about the ethical transformation, however conceived? This process of moral growth through articulation cannot be separated from Taylor’s notion of freedom, which requires that one identifies one’s more important purposes and resist or overcome desires that do not authentically reflect oneself (1985b, 213–217). In the end, the capacities relevant to establish freedom are identical to those which define our sense of motivation, in that they must involve “some self-awareness, self-understanding, moral discrimination, and self-control, otherwise their exercise could not amount to freedom in the sense of self-direction” (1985b, 215).

The implications of Taylor’s positive concept of freedom for the present discussion can hardly be underestimated, for acting in a way that actually realizes one’s self-understanding is precisely what being motivated feels like—that is, in the stronger sense of following a path toward emancipation and freedom. This phenomenology clarifies not only that our desires may frustrate our deeper purposes (as they clearly sometimes do), but also that we may be wrong or mistaken about these purposes themselves (as we clearly sometimes are). As noted at several points in the argument so far, to say that moral or spiritual values have the remarkable potential to empower human beings is to say something stronger than the widely shared philosophical assumption that values “motivate,” insofar as this concept is used uncritically; that is, to see motivation as a “relatively straightforward” and “strikingly regular and reliable phenomenon” (Rosati 2016). But if the distinction I am making between the general phenomenon of motivation and the broader, multilayered process of emancipation makes any sense, then it seems at least worth considering the possibility that Taylor’s thick moral phenomenology opens up a question about motivation that lies beyond any metaethical analysis of the nature of moral beliefs.

7 Taylor’s Answer

We are returned again to the point that there is more to the philosophy of motivation than the philosophy of beliefs and truths. If we now finally return to Taylor’s concern about adequate sources, then we might rephrase this concern as the worry that what we, late-modern Westerners, have is only a shallow moral understanding insofar as we rest content with the brute facticity of values on which robust realism insists.Footnote 12 By emphasizing the high moral demands we place on ourselves in the Western world, Taylor seems to be suggesting that we might not know exactly what we are doing or how we are doing it when we understand ourselves and others as subjects who are worthy of solidarity and benevolence. Intriguingly, robust realists explicitly refuse to allow such criticism by insisting on brute facts.

However, if there really is “no intelligible and plausible answer to [the question of] what it is in virtue of which these fundamental moral rules are true” (Shafer-Landau 2003, 96–97), then the emancipatory force of what we care about seems to evaporate altogether. Taking this position is annihilating our sense of motivation in its very meaning, closing off the entire space within which our motivations as defining orientations can be articulated and discussed.Footnote 13 This is not just intellectually unsatisfying but potentially dangerous. If robust realism’s explanatory impotence were the last word, that would make us normatively impotent as well. What we would then lack is moral understanding in the deeper sense of being capable of transforming and liberating us, an understanding on which our policies and institutions depend. As Taylor makes plain, not having (at least the prospect of) this transformational kind of understanding poses a problem to the extent that it is only this deeper knowledge that empowers us to sustain our affirmations. It is one thing to sustain the demands of solidarity simply on the feeling of undischarged obligation, or, conversely, on a sense of our own worth. But, as Taylor insists, “it is quite a different thing to be moved by a strong sense that human beings are eminently worth helping or treating with justice, a sense of their dignity or value” (1989, 515).

Significant though this point may be within a philosophical climate that is either blind to or takes for granted the emancipatory dimension of motivation, Taylor’s concern is not merely about motivational strength or the quality of our motivation. It is also, perhaps more fundamentally, about the inherent incapacity of secular sources to sustain our highest ethico-political commitments. His estimation is that even if one was authentically moved by secular moral sources (“disengaged reason” and “the goodness of nature”) as strongly as possible, this deep moral engagement would probably not be enough to sustain one’s commitments to universal justice and benevolence. On the contrary, Taylor’s hypothesis is that “wherever action for high ideals is not tempered, controlled, ultimately engulfed in an unconditional love of the beneficiaries,” these ideals may in the end only lead to suffering and destruction, giving way to an ugly dialectic where “the higher the sense of potential, the more grievously real people fall short, and the more severe the turn-around will be which is inspired by the disappointment” (2007, 697). In Taylor’s diagnosis, then, the only way out is to fully participate in God’s love or the “network of agape,” and thereby be “engulfed in an unconditional love of the beneficiaries” of our actions (2007, 739–742).

This means that Taylor’s final message concerns both the quality of our motivation and the (in)adequacy of secular sources, which brings us right back to his initial sense in Sources of the Self, that there is some good beyond mere human flourishing which would allow for “a divine affirmation of the human, more total than humans can ever attain unaided” (1989, 521). The unusually broad significance of this vision should be clear: Taylor raises the “metaethical” question of what best explains our motivation only to open up the more fundamental “normative” question of what we need, ethico-politically speaking, to accomplish what morality demands of us. At this point of the discussion, it seems a truism to say that Taylor is not concerned here with a thin metaethical question about the metaphysical status of our values without any political and cultural implication. Rather, his question is the thick and truly moral-phenomenological one of what “inspires us to embrace this morality, and the evoking of which strengthens our commitment” (Taylor 2007, 693).

This is now what an uncompromising moral realist such as Taylor has to say where there seems nothing the robust realist could say, except perhaps for a peculiar proposal to put these questions to rest. Such a proposal stifles rather than empowers our actions, for we disagree fundamentally over the sources of solidarity. This is a disagreement at the intersection of normative ethics and metaethics, and it is important. We want to have a grasp of why we should live up to the demands of solidarity and to be motivated by the good itself, not just know what the relevant facts are. Moreover, if Taylor’s diagnosis and Rosati’s observations are on point, as I think they are, then those who insist on framing moral motivation in a form that is apt for evaluation in terms of beliefs and truths are not just neglecting the emancipatory dimension of motivation. More disturbingly, they blind themselves to the vulnerability of our motivations, and the moral risk such vulnerability contains.