Affectivity in Its Relation to Memory

It seems obvious that various feelings (various kinds of affectivity) are memorized, forgotten, and recollected to various degrees. Some of them are forgotten. Some of those forgotten can be recollected, while others are lost forever. For example, short and long-lasting feelings and shallow and deep feelings are memorized and remembered in different ways. In this paper I analyse from a conceptual point of view several categories of memory-of-feelings and offer a comprehensive map of them. In the end, the richness of categories in the realm of memory is inter-preted as a proof of the intricacy of affectivity.


Introduction
When affectivity is considered in its relation to memory, two issues are at stake. The influence of affectivity on memory, i.e. emotions as modifiers of memory, 1 is not the same as the memory of emotions, i.e. the extent to which feelings 2 are subject 1 See Hillman (1960: 175): "The selection, retention, forgetting and confusion of memory representations are the result of emotion." 2 I prefer feeling to emotion, the former being broader (provided it is not understood as bodily sensation only, such as a feeling of cold, hunger etc.) and better covering the whole of affectivity, especially when affectivity is viewed as multilevel. The standard word in German is Gefühl (not Emotion). Some native English speakers have told me that I am wrong in this. I think this is an ad persona argument and though I don't possess full linguistic competence in English I time and again find evidence of feeling being used as an equivalent to emotion or as the general category for comprehending the whole of affectivity. For to forgetfulness and remembering. 3 For instance, in a highly influential handbook, The Nature of Emotion: Fundamental Questions (Ekman and Davidson 1994) one of chapters is devoted to "the Relation Between Emotion and Memory". In this chapter, one paper is about the "major influence of emotion on memory" (Bower 1994: 303), another about "how emotional memories are elicited and […] the connection between the current encounter and the past emotional memories" (Lazarus 1994: 306). Although Lazarus underscores that "[i]t is important to distinguish emotional memory from memory of emotion", since "[t]he latter is a declarative, conscious memory of an emotional experience. It is stored as a fact about an emotional episode.
[…] emotional memory and memory for emotion are stored through different brain systems […]" (LeDoux 1994: 312), memory of emotion is not treated. In what follows I shall treat the latter issue. 4 Moreover, I intend to approach it from a categorial angle. I look for a conceptual framework of memory of emotions. In such a task empirical evidence may be only of auxiliary use. I resort to such evidence when I need examples of categories suggested within the whole frame. Finally, my concern is to look at affectivity as a hierarchical or multilayered realm. This kind of approach is rare and, what is worse, is often represented by means of notions of depth or the high, both metaphorical and, for that reason, philosophically unsatisfactory. Currently I am working on an attempt to render multilayeredness of affectivity without metaphor.
It goes without saying that in what follows I do not intend to give a solid solution. I rather intend to provide room for several categories used in the debate on affectivity. In so doing I undertake a conceptual exercise and point to the taxonomical table. Since, no doubt, my results will be provisional, I prefer my umbrella to be too broad with a further need to restrict it rather than to be too narrow, in which case an item lying outside it would be hard to grasp and integrate. In fact, I believe that it is better to start with too many and then use Ockham's razor to cut off and throw back what is needless rather then to start with a lack of distinctions and risk losing any element of the whole picture-hence the idea of using something like the periodic table of the chemical elements.
For instance, in the here and now, I will speak about thoughts, feelings, emotions, memory, and so on. But I wonder if thought and feeling are separable at all. The question is whether there are such mental items as pure feeling or pure thought at class level-and the same is valid, I suppose, for pure memory, or, at the family level, pure sorrow, pure joy, etc. If not, and if feeling and thought are inseparable, as I suppose they are-a thesis I cannot defend here would be that a mental event is (conceptually) composed of several components, 5 call them thought and feeling depending on which of them is slightly or significantly predominant-a question arises as to whether they are inseparable ontically or (only) epistemically. And if the latter, will the current state of affairs change if we acquire the relevant tools to confirm their separability, or are such tools impossible, such that we are doomed forever to being unable to decide whether thought and feeling are separable only epistemically or epistemically-because-ontologically? 6

A First Distinction
With regard to memory, affective phenomena can be distinguished by means of three categories: being memorizable, being forgettable, and being recollectable. Memory may be broadly understood as what "refers to our ability to consciously reflect on past experiences" (LeDoux 1994: 311). However, I would suggest drawing a distinction between simply remembering about a past experience (say, that I had such an experience, full stop) and a full-fledged memory of what it was to feel it (say, remembering a particular quale). That some feelings are memorized is obvious, but so is forgetting and recollecting feelings. My claim is that whether they are or are not memorizable/forgettable/recollectable is a categorial criterion that distinguishes one kind of feeling from another. This distinction is to be read as intrinsic, not as accidental. It is not that the same feeling in some cases is memorizable/forgettable/recollectable and in other cases it is not, with the difference depending on, say, external factors. 7 If a feeling is memorizable rather than forgettable, this is because of its intrinsic features. If I remember well, say, some sadness that occurred 2 years ago and I don't remember another sadness that occurred a week ago this is because the former has such a feature that makes it rememberable and the latter does not. This feature marks the substantial difference between both and, at the same time, rules out the opposite; that is, the case when the former is forgotten forever and the latter is rememberable for, as mentioned, extrinsic reasons. For instance, by speaking about a past event, showing pictures or photos, or building a chain of associations I make someone remember that she had in the past a kind of feeling without, however, her feeling what this feeling was to her. In other words she will know rather than feel that she experienced such and such without remembering the specific quality of that act of feeling, i.e. without having access to the content of this such and such. I may have a memory that in my childhood I used to experience the arrival of spring in a specific way. Imagine I meet a friend after many years and he tells me about my particular feeling maybe I used to speak to him about it so frequently that he kept it in mind. At this I answer him: "yes, true, now I remember about it when you remind me of it. I remember now it was something exceptional, but I no longer remember what it was exactly". In fact, now his and my memory are of the same order even if in the past they were of different orders: mine is direct, his is second order. But that is all and I am no longer able to remember what this kind of feeling was. If this suggestion is acceptable, then the following theoretical model including three possible cases-because of the three distinct processes that occur-may be proposed. Feelings are either (a) forgettable and unrememberable = definitively lost, (b) forgettable and rememberable = retrievable, or (c) unforgettable = always presentable.
Obviously, not every retrievable feeling is retrieved and not every presentable feeling is presented. The question of when and on which conditions this occurs is beyond the scope of this paper.
Since the memory of a past feeling is not the same as a current feeling, always presentable means two things. In some cases feelings are presentable because they are current; I do not consider this category here because what is current is not yet memorized. As Aristotle puts it, "memory is of what has occurred. When what is present is present, such as this white when [one] sees [it], nobody would say [that he] remembers [it]." 8 In another sense, always presentable pertains to past feelings that have never been forgotten, which does not mean that we are aware of them all the time. Even if your attention is not directed at them ceaselessly, always presentable means that you can call them to your awareness at any time in the same manner as I can call to my awareness (not remember, properly speaking) what I did a couple of hours ago, or the postcode of my office, etc. Just as I need to think a little in order to have it present in my mind, so too do I need to feel a little in order to bring my feeling into awareness. When sleeping I am not aware of them either but this is not to say that I have forgotten them. 9 The same concerns my knowledge, say, of my name: I don't think about it all the time, yet it would be odd to say that I have forgotten it each time my awareness or attention is not directed at it.
Examples of (a), (b), and (c) can be various. Try to remember some sorrow, anger, joy, love, or fear you experienced, say, in 2010 and then think about a more recent situation when you have felt such feelings. Or look at a recent photo of yourself, or better a series of photos from the same week, and see if you can remember how you felt when the photos were taken. There are, supposedly, some feelings you don't need to recollect since they are presentable and are present, others that come to mind because of seeing a photo, but otherwise forgotten, and then those you cannot recollect even if you are sure that you were experiencing something at this or that moment.

A Second Distinction
Now, think about the duration of feelings, that is, about the duration of what is supposed to be an object of memory. From this point of view we can, I think, broadly distinguish between (i) momentary, (ii) short-lasting, and (iii) long-term affective phenomena. The distinction is not sharp and I should maybe explicate it as: (i) tending to zero duration, (ii) any middle duration. 10 (iii) tending to-but not being of-full time duration, 11 Examples: (i) A feeling being both a result of and simultaneous to, say, having a massage. You can have a shorter or a longer massage (or undergo shorter or a longer torture). A feeling lasts as long as the massage (or the torture) but normally it lasts no more than a couple of hours. This is the feeling that ends when the massage or the torture stops. This is a feeling of joy or suffering of being, respectively, massaged (provided you like being massaged) or under torture (provided you don't like being tortured).
(ii) A feeling that results from having had a massage, felt after the action inciting it has stopped. This is a feeling of joy stemming from the fact that I experienced a massage, that, say, I had longed for, I had needed or wanted to be acquainted with. This example stands in contrast with the previous one, even if the circumstance is apparently similar and for some may look like a continuation of (i). 12 But here are many other feelings too: short-term liking or disliking of a person, anger, fear lasting for a couple of days (in contrast to a sudden and short fear of, say, a wild animal), etc.
(iii) A long-lasting experience, a relationship with another person, being interested in or concerned with something, or, to draw contrast with anger under (ii), resentment. This is a feeling of happiness, i.e. a long-lasting affective experience, as when someone says: when I was a child (or a student) I felt (generally) happy.
In order to grasp the difference between (ii) and (iii), imagine a relation with a person, say a person you like [= (iii), long-lasting] and the joy you feel when you meet her after a longer time [= (ii), short, or if you prefer, shorter-lasting feeling because lasting significantly shorter than the relation itself which is the same as the long-lasting liking of that person]. And as for a distinction between (ii) and (i), this corresponds to a difference between the joy you feel when you meet after a longer time someone you have missed and the pleasure you feel when you hug her-a pleasure lasting as long as the hug. The divide would therefore be threefold and could be pictured as follows: (i) momentary, i.e. fleeting feelings, (ii) short-lasting feelings, (iii) long-lasting feelings.

A First Classification
Now I am going to combine the memory-relation feature with the time-relation feature of affectivity. It could be tempting to be swift and to assign (i) to (a), (ii) to (b), and (iii) to (c), such as to identify: This could, however, be simplistic and not in agreement with my principle of offering an umbrella as broad as possible for reviewing categories of memory-offeelings. Therefore, let me suggest a table with nine categories as follows (Table 1).
We shall leave this aside for the time being.

Going Further
Do these nine categories offer a full picture and at the same time avoid reductionism? I think not. There is, apart from duration of feelings, another factor we have to take into account. This is the level (or depth) of affective experience, in other words the level at which a feeling is experienced. This category is too often confused with intensity. 13 This is a mistake, since, for example, no matter how intensely you feel a pleasure it is never converted into joy and no matter how mildly you experience a love, it is not transformed into liking. Consequently, there are less and more intense pleasures and less and more intense joys, but the least intense joy is a different feeling to the most intense pleasure (this is what could be suggested provided that pleasure and joy are taken to be neighbouring feelings of the same genus). Here we have, to use Nicolai Hartmann's concept, 14 gaps between, e.g. pleasure and joy, joy and happiness, and happiness and blissfulness. There is no parallel with hues and what I suggest is visible in a well-known diagram of Plutchik's wheel of emotions, though I don't know if Plutchik meant to make sharp divides between three emotions within each of eight groups, e.g. between amazement, surprise, and distraction, or whether he believed the opposite and would claim that any three emotions of each group are one quality with varying degrees of intensity. 15 Let me suggest another threefold divide, which expands on the previous one by adding shallow feelings, middle-depth feelings, 16 and deep feelings ( Table 2). If I adopt three categories, as in the previous case of duration, I do it for the sake of simplicity. Two are not enough because two dichotomizes the perspective, but three are sufficient to set a balanced framework. If, however, there are actually four or five or more, the picture would be different but not so much as to affect the threefold model in its principal elements.
I am well aware that the concept of depth is metaphorical (to the same extent, I suppose, as high and low). This concept is applied both in non-scientific and scientific, non-philosophical and philosophical discourse, starting with Heraclitus. 17 Max Scheler used it in his fourfold model of affectivity 18 but neither he nor anyone else I could find has been able to translate it into non-metaphorical concept at all, let alone in a satisfactory way. It must, it seems, be understood intuitively. But how exactly?
I would tend to explain it as a degree of submersion of the subject into the affective process, more precisely the part of the subject involved in affective experience. We may think about a person being completely submersed in a computer game. But in this case I would say that, although entirely submersed, the submersion is shallow, for it is not a person's whole existence that is concerned unless (and this might not be implausible) her existence is very poor and flat. Accordingly, as proposed by Scheler (1973: 332-344), experiencing a feeling localized in a specific part of the body is not same as one engulfing the whole of one's self. Because one could 14 See Hartmann's "law of distance between levels" (in: Hartmann 1953: 76). 15 Plutchik (1991: 109) mentions intensity: "In order to develop this analogy [i.e. between color-sensations and emotions], it is necessary to conceive of the primary emotions as hues which may vary in degree of intermixture (saturation) as well as intensity, and as arrangeable around an emotion-circle similar to a color-wheel." But if so, the representation of his wheel is incorrect since there are clear-cut borders between hues instead of fluidity within each of 8 emotion groups and, that being so, intensity is not represented. 16 An intuitive reason for introducing the middle-depth level is to avoid dichotomisation. But there is more: while one may think about extreme cases of duration, for example a short or tending to zero duration and another one tending to everlasting, something in between, tending to neither, seems to be conceptually required too, at least at this stage of the analysis. I recognize that this take is Aristotelian. 17 Heraclitus (1989, DK 22 B 45): βαθὺν [λόγον]. 18 Note that he uses italics with inverted commas to mention it (e.g. Scheler 1973: 332), or simply italics (e.g. Scheler 1973: 364), or neither (e.g. Scheler 1973: 349). think-I happen to think-that an example of a shallow and an example of shortlasting feeling are the same example, e.g. a massage, as mentioned above, it could be tempting to conclude that the shallower an experience is, the shorter it is and the deeper it is the longer it lasts and, next, to identify momentary feelings with shallow feelings, short-lasting feelings with middle-depth feelings, and long-term feelings with deep feelings. This is often the case, but it is not necessarily so, as this happens in the case of an extremely disappointing/captivating affective experience lasting a very short time that is very well remembered even if temporarily forgotten and then retrieved. 19 For instance a short-lasting hug and the short-lasting feeling that accompanies it can be as deep as to penetrate into the category of being temporarily lost (retrievable) or even unforgettable. And the opposite is not always true either: a long-lasting affective experience can be forgotten after many years, for example it occurs with feelings that last or recur throughout childhood that are then forgotten forever. 20 I cannot remember feelings I experienced for years every morning when leaving for school though I now know there was something peculiar about this moment during a period of my life. Examples from Thomas Mann's novel Disillusionment (1896) are particularly insightful for proving that the opposite is also possible: I remember, sir, with painful clearness the first disappointment of my life […] There was a fire at night in my parents' house, when I was hardly more than a child.
[…] I discovered it first, and I remember that I went rushing through the house shouting over and over: 'Fire, fire!' I know exactly what I said and what feeling underlay the words, though at the time it could scarcely have come to the surface of my consciousness. 'So this,' I thought, 'is a fire. This is what it is like to have the house on fire. Is this all there is to it?' (Mann 1936, transl. H. T. Lowe-Porter) 21

Reconsideration
I have arrived at the end. I have taken into account three perspectives with a total number of 27 affective types in view of memory. The total of 27 results from combining a linear feature of memory (three categories: definitively lost, retrievable, and unforgettable) with a linear feature of duration of spontaneity (three categories: momentary, short-lasting, long-lasting) and with a vertical perspective (levels, or depth) of affectivity (three categories: shallow, middle-depth, deep). Now after thus far being Mendeleevian, it is time to get hold of Ockham's razor and make cuts. This is because in Table 2 out of 27 categories of memory-of-affectivity some may be empty and, indeed-as we shall see-some turn out to be empty. But before that, let me say that the model in Table 2 is not necessarily a full one and perhaps could be developed further. I have in mind a hint offered by J. H. Jackson in his Croonian Lectures. Jackson's approach is empirical, so I cannot consider it here, but if he is right 22 then it could be necessary to envisage a further subdivision of levels into It can be objected that his short-lasting experience was memorized by him because it was strong enough to be memorized, this, in turn, because "[he] got some burns". I don't however mean his memory of the fire but his memory of being disappointed and "feeling [that] underlay the words". 22 As far as I can see he may be right: think about the class of experiences of senses of which some are better memorized than others: views and sounds much better than tastes and smells. When I say much better I rely on the fact that views and sounds are easier to reproduce or imagine than tastes and smells. I cannot develop this claim here. See Ribot (1903: 142) on "a provoked recollection" in the case of tastes and smells.
subdegrees or subdepths 23 in view of memory-of-affectivity. I leave this point aside and proceed to eliminate virtual categories of memory-of-affectivity. The task of eliminating them could be carried out empirically, I suppose, but here I limit myself to a conceptual task. The first distinction to be made is between those that can be exemplified (as the two categories exemplified above) and those that cannot. And the latter occurs either because they are ontically empty or because they are epistemically empty, this is we cannot find any example of them and think of them as empty, while in fact they could be exemplified. For the sake of description let me call the former manifestly empty categories and the latter probably empty categories. I shall apply bearing in mind that my table is a kind of Mendeleevian model, avec toutes proportions gardées, and all modifications, additions, and removals are welcome. In this sense the model I arrive at below is a first attempt at offering a map of categories of memory-of-affectivity.
I think that (7 I ), (8 I ), and (9 I ) are manifestly empty because shallowness of experience flies in the face of being memorized forever; if not, my classification is wrongly conceived altogether. For the same reason I would rule out (4 I ) and (5 I ).
As probably empty I would suggest: • (6 I , shallow retrievable long-term feelings) because the shallowness reinforced by a long duration could perhaps result in retrievability of memory of feeling.
Probably empty are also: • (3 I , shallow definitively lost long-term feelings) and • (2 I , shallow definitively lost short-term feelings) because it is not certain-otherwise examples are to be offered-that shallowness can go together with the short-and long-term duration of a feeling, whereas (1 I , shallow definitively lost momentary feelings) seems to be existent beyond any doubt as exemplified by the everyday experience.
1 3 Axiomathes (2018) 28:253-267 As for the third group-deep feelings-(7 III , deep unforgettable momentary feelings), (8 III , deep unforgettable short-term feelings), and (both 6 III , deep retrievable long-term feelings and 9 III , deep unforgettable long-term feelings) are existent categories, which comes from the way depth is understood: if depth, regardless of the duration of a feeling, is not what warrants the unforgettability of a feeling, then the whole scheme is flawed. 24 If depth is a warrant for the unforgettability of a feeling, then it can be inferred that there is no such thing as deep and definitively lost feelings (regardless of their duration): hence (1 III , deep definitively lost momentary feelings), (2 III , deep retrievable short-term feelings), and (3 III , deep retrievable long-term feelings) can be considered manifestly empty. And (4 III , deep retrievable momentary feelings), and (5 III , deep retrievable short-term feelings)-deep retrievable feelings if momentary of short-term-are probably empty, at least as long as no counterexample is provided. If one is provided, we should also enquire whether duration interferes within this group classification as a criterion.
The most disputable is the second group: middle-depth feelings. The difficulty is maybe due to the general characteristic of in-between cases. Here I proceed intuitively and schematically by linear steps and degrees as follows: • for unforgettable middle-depth feelings, probably empty categories are (7 II , middle-depth unforgettable momentary feelings) and (8 II , middle-depth unforgettable short-term feelings), while (9 II , middle-depth unforgettable long-term feelings), given its duration, is existent (hence there are no manifestly empty categories in this subgroup), • for retrievable middle-depth feelings, category (5 II , middle-depth retrievable short-term feelings) is probably empty, while (4 II , middle-depth retrievable momentary feelings) is manifestly empty, and (6 II , middle-depth retrievable longterm feelings) is existent, • for definitively lost middle-depth feelings manifestly empty categories are (1 II , middle-depth definitively lost momentary feelings) and (2 II , middle-depth definitively short-term feelings), while (3 II , middle-depth definitively long-term feelings) is probably empty (hence there is no existent category in this subgroup).
If the above provisional arrangement makes sense it could be presented as follows (Table 3). 25 Some further observations can be made. Depth is a stronger criterion than duration. This is visible in I vs. III. In III the distinction is horizontal because depth is decisive and duration plays no role, while in I duration is crucial for distinctions. Finally, II, though in between I and III, is not so much a simple mixture of I and III, as one could mix some elements of both. For example given that in III depth is a prevalent criterion and in I depth is modified by duration, one could ask to what extent duration is a modifier of depth degree in II. There is also another peculiar property, namely that antipodes are characterized by specific features, in I 5 out of 9 are manifestly empty, while in III 4 out of 9 are existent. From now on, more examples and descriptions are needed for each particular category. One should be careful to avoid biased examples as well as to be attentive to their relevance to representativeness, unless one sets forward an argument for disapproving the whole idea of the table thus conceived. Now, if one wishes to throw out manifestly empty and probably empty categories-though I am not keen on doing so, still having in mind the above-mentioned principle of the broadest umbrella possible, and, thereby, would do so only for the sake of transparency of the hypothetical result-the result would be as follows ( or unforgettable [if middle-depth or deep: (9 II ) and (9 III )], • deep feelings are always unforgettable (7 III ), (8 III ), and (9 III ), • short term feelings are unforgettable only if they are deep (7 III ).
And the negative conclusions are that: • there is no short-term (2 I ), (2 II ), (2 III ) nor long-term feelings (3 I ), (3 II ), (3 III ) definitively lost, 25 The model should, it seems to me, be analysed in two separate ways: for active memory on the one hand and passive memory on the other. This is because in some cases feelings are remembered in the sense of recognition when a similar phenomenon comes about. I remember because I see, hear, read etc. something similar (or dissimilar) that provides me with a clue and by way of association a recollection. I would call this a passive memory of feelings. In other cases feelings are remembered or recollected without such a supporting element. This would be an active memory of feelings. Compare Stein (1989: 8-9): "This process can occur passively "in me" or I can do it actively step by step. I can even carry out the passive, as well as the active course of memory without reflecting, without having the present "I," the subject of the act of memory, before me in any way. Or I can expressly set myself back to that time in a continuous stream of experience, allowing the past experiential sequence to reawaken, living in the remembered experience instead of turning to it as an object." • there is no momentary (4 I ), (4 II ), (4 III ) nor short-term feelings (5 I ), (5 II ), (5 III ) retrievable. 26 Finally, I is represented by one category only, II by two categories (of the same subgroup of long-term feelings), and III by four categories (of the same subgroup of unforgettable feelings plus one of retrievable feelings).

A Tentative Conclusion
In my paper I have taken a conceptual path. I wanted to draw categorial distinctions pertaining to the way various kinds of emotions are remembered. To this end I have tried to shed light on some basic distinctions relevant to the memory of affectivity. First I brought under consideration as many categories as seemed possible. Next I proceeded to a conceptual reduction. I presented my framework of categories as hypothetical and tentative rather than a firm and definitive taxonomy. The proposal is to be tested but what looks certain to me is that the realm of memory of affectivity is another field in which the intricacy of affectivity 27 is manifest. Its intricacy lets us admit that without being homogeneous affectivity is still one class or family of mental phenomena. From this angle my framework can contribute, I hope, to a multidimensional model of affectivity, more promising by itself than a one-dimensional model, which too often ends in contradictions and/or leads current debates on affectivity to insolvable aporias. I acknowledge that my rationale for distinctions looks at some points circular. I explain retrievability as determined by depth of experience and depth of experience as what determines the retrievability of a feeling. This is a manifest circularity at the explanatory level. But what if the two features-retrievability and depth of experience-are inseparably related? In that case I have-at present at least-no answer other than the one given by B. W. Helm when defining the emotions by means of import: "import and the emotions emerge together as a holistic package all of which must be in place for any of it to be intelligible. The circularity of the account is therefore a normal part of such holism and is not at all vicious" (Helm 2010: 313). Likewise, depth or what depth stands for and retrievability, together with constant presentability, are inseparable characteristics.