I’ll start with a personal moment of propatheia — Stoicism’s term for a knee-jerk reaction to something that one experiences in the world. I was making a left turn the other day during rush hour. As I moved into the intersection, a pedestrian was crossing at what I thought was a rather leisurely pace. My immediate response was to wonder why someone who could easily move faster not make a more brisk attempt to cross quickly and keep the flow of traffic moving. I might call this a knee-jerk feeling that could only be described retrospectively as annoyance, but certainly not anything like anger. I’ve been trying to work on the habit of observing these moments of judgment and turning my attention away from the actor so as to make sense of my knee-jerk reaction. Why did I have this reaction? Was it fair to the person crossing the street? I could not have had any idea as to her motivations, mental state or physical capabilities.

Of course my knee-jerk judgment wasn’t fair. Yet I had easily slid into a common mental model of judgment and moral responsibility: I’m the judge who renders another’s actions worthy of praise or blame. Central to this model is the belief that this person crossing the street had some level of autonomy over her actions — call it “free will,” which must exist at some level for this mode of judgment to work. She was freely choosing to ignore the world rushing around her and meander across the street — or so the judgment goes. This mental model focuses all of the moral responsibility on her by making her choice into the object of judgment. In other words, I am the judge of her judgment. I have moral responsibility in this situation only if I have actual authority over her to either change her current behavior or exact punishment (or correction) for her past behavior. Without that authority, I’m merely an observer of the situation.

For my judgment to be valid, however, it needs to be considered neutral and objective — i.e., unbiased and free from any subjective, personal influence. Moral philosophers have invented various rules to enforce my neutrality and objectivity as the observer of moral rights and wrongs. I might use one of Kant’s categorical imperatives; I might rely on Utilitarianism; or I might look to the Golden Rule for the ground of my objectivity. Using any of these criteria, I am an absent presence as the judge. I’m present insofar as I am the consciousness that renders the judgement; I’m absent in the sense that I’m trying to set aside any personal biases so as to rely on purportedly objective rules that are theoretically accessible to anyone, anywhere at any time. Thus the objectivity is simultaneously a neutrality — it must be unbiased and non-subjective — and universal — the rules that I apply must be true for all situations that resemble the current one under consideration.

In this meditation, I want to take seriously my self-awareness that happened in this episode. By turning my attention to my own reaction, I interrupted the knee-jerk model of judgment and activated a different model of judgment — one that starts by judging my response to someone else’s action. This model is, I believe, at the heart of Seneca’s vision of moral responsibility, and I want to use this and a few subsequent meditations to more thoroughly explore its implications. In these meditations, I’m going to borrow some concepts from Henri Bergson as a way to hold attention on the temporal dimension of moral responsibility; that is, I want to explore the implications for moral responsibility when we take seriously that human actions and awareness of them unfold in a flow of time (Bergson’s duration) rather than being driven by stable entities like reason, emotion and free will.

I do believe that a pairing of Bergson and Seneca is appropriate and interesting for a couple of reasons. One is that Seneca’s therapeutic approach to philosophy is accessible and readily understandable to the modern reader, but that accessibility makes it easy to gloss over some of the challenging implications of his writings. We too easily find a Kantian approach to moral judgment (i.e., I judge the actions of another using universal criteria from a position of neutrality). We too easily ignore the fundamental role of feelings and emotion as the provocations to reason and action. We too easily find a coherent “self” as the owner and activator of a free will. And we too easily see him as offering a concept of the mind as the home of reason and passion as separate and distinct psychological entities. To be sure, all of these are legitimate readings of Seneca. They aren’t completely wrong, and I’m not trying to get Seneca right. Rather, I want to spend some time taking seriously his emphasis of temporal metaphors to describe what is happening to the mind as it responds to a world experienced as a string of unpredictable events.

The other reason for pairing Bergson and Seneca is that Bergson’s metaphysical approach, while not nearly as accessible to the modern reader, does provide some concepts that allow me to hold my attention on the temporal dimension of Seneca’s writings. Intensity, duration, and qualitative multiplicity can be interesting lenses through which to pull out some important implications for moral responsibility, will, determinism and experience that are certainly “in” Seneca’s text but are easily lost if we read him as primarily a metaphysician, or if we read him using our common sense lenses bequeathed to us by moral philosophies driven by universal criteria and rules. Bergsonism helps me suspend the spatial and universal bias so that the temporal can come to the fore and hold sway over my reading of On Anger.

Let us begin by remembering what is fundamental to Stoicism and Seneca: we are only responsible for what we can control, and all that we can control is how we respond to the unpredictability of what life throws at us. While all events are governed by providence (providentia) as a tightly woven web of cause and effect, the human mind/soul (animus) has no access to the complete knowledge that would make this cause-and-effect transparent or foreseeable. Therefore the human being experiences the highly determined as unpredictable. This paradox is inherent in his use of the term fortune (fortuna), and it is fundamental to Seneca’s thought. To put it in the Latin terms used by Seneca: Providentia is experienced as fortuna.

This fundamental paradox is the inescapable human condition for Seneca. This, I believe, is why he sees no reason to agonize over our modern philosophical obsession to reconcile free will with determinism. Though plenty of scholars have pointed out that this tension is fundamental to Stoicism, Seneca clearly doesn’t see it as a philosophical problem that needs to be solved. Nowhere do we find anything like “free will versus determinism” as a topic of discussion in Seneca’s work.

If we take this proposition seriously, Seneca’s writing does three things that are important for asserting moral responsibility in a determinist’s universe. First, Seneca makes time more important than space in describing mental events. The importance of this temporal bias cannot be underestimated when reading Seneca. I’ve already meditated on how Seneca’s concept of reason is a time-based concept, not a spatial one rooted in a faculty of the mind. I also recently read Brad Inwood’s arguments against finding “dualism” in Seneca where he emphasizes human experience as fundamentally time bound:

The soul he describes is unitary; the complexity which Seneca introduces centers not on parts or powers, but rather on a temporally dynamic process (Reading Seneca, 55; Inwood’s emphasis)

Second, his bias toward time offers a different sequencing of judgment, action and responsibility. Crucially, judgment happens twice in the sequence: (a) one has a feeling (propatheia, or “pre-emotion”) of being wronged by another and this leads to a knee-jerk judgement of the other, which is what I did in the busy intersection; (b) before we act, we must turn the judgment back on ourselves to understand and evaluate how we are responding. (Inwood has discussed this at length in “Judgment in Seneca,” and it is readily on display throughout On Anger.) Responsibility, as I’ll take up later in this mediation, happens at (b) and is about our response to another’s actions. Responsibility, in other words, is not merely about the actions of the actor. It is about what we can control, which is our response to another’s action. For Seneca, moral responsibility happens as these transformations unfold in the respondent as she reacts to the situation at hand — a situation that is simultaneously highly determined and unpredictable for the human mind. Again, providentia is experienced as fortuna.

Third, self-control is about governing the speed of our responses, and we exercise self-control by becoming aware of the transformations of these responses. Improving one’s ability to become aware of these responses at morally significant moments is freedom for Seneca. Thus self-awareness is crucial to moral responsibility, and it is fundamental to Stoic concepts of self-examination, training, joy, freedom and the ability to endure hardship — all of which are ways of dealing with unpredictable events in the flow of time.

None of this requires a philosophical commitment to anything like “free will.” Free will is not something we find in Seneca, though we do find “freedom” (libertas) and “will” (voluntas) as prominent concepts in his worldview. Both are related to conscious action, but they are decoupled such that “will” is an effort exercised in the flow of time, but it is not free in the sense of self-starting, nor does it have an origin as a pyschological entity. Freedom (libertas) is a state of consciousness that is the result of the exercise of will and not its cause. Seneca is not a proto-Existentialist; we are not “condemned to be free” as an original condition of human existence. Seneca’s freedom, joy (gaudium) and tranquility (tranquilitas) are achieved by training and practice in making us better at becoming aware of how and why we react to life’s unpredictable events. This awareness is the starting point for self-control exercised in the flow of time rather than the activation of psychological entities. I shall return to this later in a close reading of some of Seneca’s more telling passages in On Anger. For the moment, it is enough to call out his bias toward formulating moral responsibility as a problem of time rather than space and that “free will” (and therefore determinism) is not at all important or present in Seneca’s thinking about moral responsibility.

To be sure, Seneca at times slips into dualism — presenting “reason” and “passion” as separate psychological entities that compete with each other for dominance in governing one’s actions. But, as Inwood has convincingly pointed out, this is a literary convention; a closer reading of Seneca always finds the temporal more important than the spatial (Reading Seneca, “Dualism in Seneca”).

Another way to look at this is through the lens of Bergson’s “intuition as method.” Insofar as any one of us tries to explain our actions, we are locked into language, which automatically introduces a spatial bias. This is what we find when Seneca engages in descriptions of reason and passion that make them into entities. But Seneca typically reverts to trying to capture a temporal experience and mitigates, as much as possible, the spatial descriptions in favor of describing temporal “transformations” in the mind. This is precisely what Bergson attempts to do with “duration” and “intuition” as concept and method respectively. Duration is the concept he uses to emphasize the human experience of time, where intuition is the methodological approach that recognizes that any attempt to explain duration is bound to fail because the explanation will invariably impose a spatial framework on temporal experience:

But we forget that states of consciousness are processes and not things; that if we denote them each by a single word, it is for the convenience of language; that they are alive and constantly changing; that, in consequence, it is impossible to cut off a moment of them without making them poorer by the loss of some impression, and thus altering their quality. (Time and Free Will, 196)

I believe that Seneca is trying to do the same thing (from a first-century elite Roman perspective) when he oscillates between his spatial metaphors and his therapeutic intent, especially in On Anger.

If we’re looking to translate the Stoic moral psychology into an ontology of the mind, we miss something crucial and practical in Seneca’s work. We ignore its fundamentally therapeutic intent in order to mine it for ontological concepts of “the self” and the related psychological entities of reason, judgment, passions, and will. This way of reading Seneca will expose contradictions in his writing and leave us with a very unsatisfactory view of him as a philosopher — witness the accusations of “eclecticism” that have dogged his evaluation by modern scholars.

When we take the therapeutic intent of Seneca’s work seriously, a more Bergsonian focus on time becomes readily apparent. Reading Seneca as a philosopher-therapist, we find him much more concerned to describe a flow of consciousness that he fundamentally believes can be controlled. On Anger, in my reading of it (even before I read Bergson), is a prolonged attempt to describe the temporal sequence of moral responsibility and not resolve it to spatial entities of reason vs passion as the fundamental ground of an ontology. I could pick random passages to show how he does this, but let me take the crucial early paragraphs of Book 2 as illustrative of this intent.

In 2.3-4, all the important terms of the Stoic moral psychology are marshaled to describe a temporal sequence. In fact, the sequence is as important as the concepts:

We [Stoics] hold that anger ventures nothing on its own but acts only with the mind’s approval: for (a) having the impression that one has been done a wrong, (b) desiring to take vengeance for it, and then (c) combining both in the judgement that one ought not to have been harmed and that one ought to be avenged — none of this is proper to a mere impulse set in motion independent of our will.

The first sentence in this passage clearly defines “anger” (ira) as an entity — it is the subject of the sentence, and it “ventures” and “acts.” Such a reading leads us to dwell on the subsequent important nouns — impression, judgment, impulse and will — as ontological entities. In doing so, we necessarily discount the temporal, and thus the therapeutic, intent of describing a sequence, which is the force of this passage. In fact, we’ll find later passages that describe this flow in slightly different terms. But the differences shouldn’t diminish the therapeutic value of the work, which is attempting to give us the conceptual tools to help us respond to the random events of life as they unfold in time — and crucially cannot be fully understood as they unfold.

This passage demonstrates a good example of intuition as method. Seneca is using spatial metaphors to describe an experience that is unfolding in time, and his text struggles to subordinate the spatial to the temporal. Let’s take the next passage to illustrate:

“What is the point of this inquiry?” you ask. The point is to know what anger is; for if it comes into being against our will, it will never yield to reason—indeed, any movements that occur independent of our will cannot be overcome or avoided…

From here he goes on to describe in detail the Stoic concept of pre-emotions (propatehia) as unavoidable in the sequence of human reaction, to which I will return. But the important point that I want to make in this passage is that the “what is anger” question is resolved as an issue of time, not a issue of entities (and thus space): the “what anger is” statement is immediately followed by a discussion of how “it comes into being”. The “will” is emphatically not an entity acting independently; it is the temporal ability to control when (or if) this “coming into being” will occur. The use of the term “movements” by the translator captures this ambiguity inherent in “intuition as method.” That metaphor nicely captures the challenge of representing time as space. What actually is “moving” in this passage? None of the so-called entities in this passage can be said to be moving on their own and colliding with each other as they vie for dominant position. (Seneca makes fun of this way of thinking in Letter 113.) Rather, “movement” here is better thought of as transformations in one’s state of mind. In other words, we’re not talking about movements of propatheia, judgment and anger; we’re talking about movements through these different states of mind. The seams between the key concepts are the necessary artifacts of the explanatory process — a process that is fundamentally therapeutic for Seneca and not ontological.

It should be clear in this sequence that complete knowledge of the situation is not fully present or available without some effort on the part of the one who is responding to the perceived wrongdoing. This is why, at 2.29, Seneca will famously say that “the great cure for anger is delay.” Earlier, however, we find the clearest explanation of controlling the pace at which one moves from impression to reaction:

… it is what follows that must be set in order. Accordingly, we must struggle against the passions’ first causes. The cause of anger is a belief that one has been wronged, to which one ought not lightly to give credence. One shouldn’t immediately assent even to what is clear and obvious, for some things are false that look like the truth. One must always take one’s time: the passage of time makes the truth plain. (2.22.1-2; emphasis added)

I will return to this passage later, but at this point I want to emphasize that this is clearly a therapeutic sequence that must wait, must delay, in order to arrive at the truth of the situation and what the proper action should be. Full understanding only arrives, if it ever does, in its own time and on its own schedule.

What exactly is being delayed? To answer this, we must return to propatheia — the “pre-emotions” that fall short of being full-blown passions like anger. They represent the pure possibility of passionate response without being determinants of the response. Propatheia are themselves describable as hard determinism. Even the sage has no control over them: “These are all movements of minds stirred despite themselves; they’re not passions but the first prelude to passion” (2.2.5, emphasis added). For them to become passions, we have to let them become so — either through positive effort or neglecting to delay the flow of mental movements. Thus anger, as an emotional response, remains only a possibility at the moment of propatheia.

(A side note on terminology: In Stoicism and Emotion, Margaret Graver points out that Seneca doesn’t actually use the term propatheia, though he is clearly embracing this canonical Stoic concept throughout Book 2. I’m following her reading of Seneca accordingly.)

If we try to treat Seneca’s moral psychology as a defined sequence of compartmentalized entities (or as discrete moments in time), we further bury the therapeutic intent of what he’s arguing. Here I find Bergson’s concept of “duration” a helpful concept to keep our focus on the temporality of what Seneca is describing. We are much better served if we think that Seneca is attempting to describe, however imperfectly, an intervention into a flow rather than compartmentalized moments of a self-reflecting individual. Bergson in fact outlines a similar moral psychology in Time and Free Will that makes use of parallel concepts of “idea” and “effort” and “act” as a temporal flow:

And from the idea to the effort, from the effort to the act, the progress has been so continuous that we cannot say where the idea and the effort end, and where the act begins. Hence we see that in a certain sense we may still say here that the future was prefigured in the present; but it must be added that this prefiguring is very imperfect, since the future action of which we have the present idea is conceived as realizable but not as realized, and since, even when we plan the effort necessary to accomplish it, we feel that there is still time to stop. (211)

For Bergson, this flow can only be segmented in retrospect as an act of evaluation; but as a lived experience, the flow can be manipulated and slowed down or even stopped so that pre-figured actions don’t necessarily follow a predetermined causal chain from idea to effort to action.

So it is for Seneca. Propatheia (equivalent in this sequence to Bergson’s “idea” as “realizable but not as realized”) does not necessarily and definitely cause the impulse that leads to the action. The possibility of passion-driven action is inherent in propetheia (sometimes translated as the “initial impression”), but it does not emerge fully formed from the cause. All that is caused in a hard determinist way is the initial involuntary reaction — the “feeling” of being insulted or injured or otherwise treated unjustly. This is something that even the sage experiences. (See Graver’s discussion of this aspect of propatheia in On Anger in Stoicism and Emotion.)

To finish this meditation, I want to dwell a bit on the importance of putting words to feelings that is fundamental to Seneca’s temporal moral psychology that I’ve been describing with a little help from Bergson. Lekta is the Latin term that encompasses this ability to put one’s propatheia into words. Lekta is how Seneca forms propositional content to the impression of being wronged. For Stoics, this propositional content is what one assents to (or revises before assenting). This propositional content takes the form of “I believe that I have been wronged (and I should do something about it)” not simply “I have been wronged (and I should do something about it).” To return to a passage already cited (2.22.2), we see that belief is the operative term, and it is contained in the propatheia:

The cause of anger is a belief that one has been wronged, to which one ought not lightly give credence. One shouldn’t immediately assent even to what is clear and obvious, for some things are false that look like the truth. One must always take one’s time: the passage of time makes the truth plain (2.22.2; emphasis added)

This is the passage that kicks off the therapeutic section of Book 2, after Seneca’s discussion about how to raise children so that they are not habituated and pre-disposed to angry responses. At 2.22.1, Seneca is explicitly talking about our adult responses to perceived wrongs. Children have no moral responsibility because they are incapable of exercising lekta. Thus the adults are responsible for raising them correctly. For adults, the moment of moral responsibility starts with the “I believe” part of the proposition. The content of the belief is what must be evaluated before one can assent to the proposition and take the desired action.

Moral responsibility thus starts, not in the evaluation of the alleged wrongful act itself, but in consciously verbalizing the content of the propetheia as a provoked belief. It is to this lekta that one reacts. Lekta poses the question, “Am I having the correct assessment of the experience?” Even if I am, the response should not be automatic and (if one is dealing with the breaking of a law) formulaic. Moral responsibility is thus experienced temporally over (a) an impression, (b) a verbal formulation of the content of the impression, (c) a judgment of the basis for one’s response to the impression and then (d) delaying and deliberating upon the actions that will yield the best outcome:

The great cure for anger is delay. Ask it, at the outset, not to forgive, but to deliberate: its first assaults do the damage, but if it waits it will back off. (29.1)

Self-control is thus a function of controlling the speed at which the moral psychology operates. Formulating our initial impression (propatheia) as propositional content (asking yourself why you believe that you’ve been wronged) is itself an act of controlling the speed of the process. Merely asking the question has the benefit of slowing things down and giving ourselves time to deliberate. Anger will naturally dissipate in this delay. This is what remains so powerful and relevant about On Anger today: it is a therapeutic discussion of anger rather than an attempt to reduce anger to an ontology or to a psychological dualism of reason versus passion.

What doesn’t necessarily emerge in the act of delay is neutrality. Rather, a reflection on one’s own shortcomings and sympathy for the human condition should intervene. In the popular disposition to moral responsibility, we typically start from a claimed position of moral neutrality, as if we are objective judges of the situation with no personal baggage. This is what happened when I judged the slowly moving pedestrian before I turned my awareness back to my own response. But Seneca is emphatic that this neutral, objective and universal position doesn't exist for anyone other than the sage (which is admittedly very rare in human history):

We’ll become more self-controlled if we take a look at ourselves and ask: “It’s surely not the case that I’ve done nothing like that myself, is it? Surely I’ve gone astray that same way, haven’t I? Is it in my own interest to condemn such behavior? (2.28.8)

Here is a very different use of lekta than simply forming propositional content as we would find in canonical Stoicism. These statements are much more self-reflective and self-searching than “I believe I have been wronged and should do something about it.” In fact, Seneca is subtly pointing out that lekta as propositional content is not good enough to effect the self-control necessary to act ethically in any given situation. One must go further in the use of language to verbalize, understand and control our responses.

In doing so, he makes self-control into a function of the myriad conceptual frameworks available to us — whether we want to call them speech acts, discursive practices, ideologies, language games or any other way in which modernity has theorized the function and power of language to create reality. I can use Marxism or feminism or Freudian psychoanalysis or Begsonian duration or Deleuzian flow or Jungian archetypes or Foucauldian power analytics or any other available language game at my disposal to shed light on my motivations and the outcomes I would like to effect.

To summarize my reading of Seneca, the locus of moral responsibility originates in the act of judgment itself, and therefore more properly can be seen as originating with the judge as he/she responds to an impression of wrongdoing by an Other. Moral responsibility thus takes seriously responsibility as response. Stepping back and verbalizing our own initial impressions and feelings (propatheia) is when moral responsibility starts. This completely undermines any sense that one’s judgements about others’ morally relevant behaviors are objective or neutral. By starting with our own responses, we shift the morally significant moment to focus on two things:

1. What is our own personal history driving our reactions? (The answer will never be purely personal but will involve an understanding of how history has led to and governed your reaction. There is a lot to unpack here, and I will have to take this up later.)

2. How do we want the situation to turn out? What is the truly “useful” reaction to this situation? (Anger, famously for Seneca, is never useful.)

This shifts the concept of moral responsibility from one founded on space — who is the owner/maker of the choice that needs to be praised or blamed — to a temporal framework — slow down and reflect on your own reaction before you do something you’ll regret. To slow down one’s reactions is hopefully to slow down the whole situation so that proper outcomes can be evaluated. Recognizing the initial moment of judgment as both determined but not causal of subsequent actions (“realizable but not realized”) is when moral responsibility occurs.

Slowing down the process rather than seeking the origin of the moral choice in the Other allows us to sidestep problems of “determinism versus free will” when it comes to assigning moral responsibility. It is unnecessary to claim any absolute locus of free will in this process just as it is unnecessary to claim any need for stable, coherent selves and subjects in the process. All we have is the flow of experience through time and our ability to control the pace of our reactions. Both Bergson and Seneca theorized this in their own ways and to their own purposes. In the time between an impression and the actions we take, we can almost always find a way to delay, verbalize and evaluate: “the effect will no longer be given in the cause. It will be there only in the state of pure possibility and a vague idea which perhaps will not be followed by the corresponding action” (Free Will and Time, 212; emphasis added).

It is admittedly tempting to turn attention to where (spatially) one’s capacity to slow down and self-reflect resides. To succumb to this temptation is to necessarily ground one’s morality on a metaphysics (or a metapsychology) that looks for ontological entities as the main actors, which I am not inclined to do despite my opportunistic use of some of Bergson’s metaphysical concepts in this meditation. One does not need a fully baked metapsychology to find usefulness in Seneca’s advice or in Bergson’s metaphysics. One does need, minimally, a cultivated propensity for self-control and the ability to dictate as much as is practical the pace of one’s judgment and the self-reflective initial direction of its attention. For Seneca, this was always the proper use of philosophy.