Hume's theory of the passions appears in Book II of his Treatise (1739), and, in shorter form, in his "Dissertation on the Passions" originally from Four dissertations (1757).(1) When the "Dissertation" first appeared, two reviews criticized Hume's theory for being unoriginal. The first appearing review, which was in the Literary Magazine, says of the "Dissertation" that "we do not perceive any thing new. This we should not mention if we were not talking of an author fond of novelty."(2) The Critical Review opens its account noting similarly that "in our opinion [the "Dissertation"] contains nothing new or entertaining on the occasion." After presenting excerpts, the reviewer concludes, "This whole dissertation, to say the truth, appears to us very trite and superficial; and unworthy of so eminent a writer. But no authors are always equal to themselves...."(3)
Hume, by contrast, believed that his account of the indirect passions, which dominates both Book II and the "Dissertation," was unique.(4) William Rose, in his review of the "Dissertation" in the Monthly Review, agrees with Hume noting that "His theory depends entirely on the double relations of sentiments and ideas, and the mutual assistance which these relations lend to each other. What he says upon the subject, is extremely ingenious, and deserves the philosophical reader's attentive perusal."(5)
Few today would dispute the uniqueness of Hume's account of the indirect
passions. Why, though, did two reviewers consider Hume's theory unoriginal?
Part of the answer rests in the degree to which Hume's general classification
of the passions follows traditional classifications. By understanding how
Hume was in line with tradition, his uniqueness may be better appreciated.
I will first sketch the traditional classifications of the passions. Second,
I will draw comparisons between aspects of Hume's classification and other
Enlightenment accounts. Finally, I will criticize the interpretations of
Hume's classification offered by Kemp Smith, Ardal, and Loeb, and suggest
what I believe is the most textually sound classification. In each of these
sections, my focus is on taxonomy. However, since many divisions of the
passions derive from the psychological mechanisms which produce various
passions, some account will be given of these operations.
The passions have been traditionally understood as disturbances of the mind or soul, the key ones being love, hate, joy, grief, hope, fear, desire, and aversion.(6) At the risk of oversimplification, a fairly unified account of the passions can be found, beginning with Plato, and lasting until the mid 19th century. One feature of this tradition is that the passions fall into positive and negative classes. This is first noted by Aristotle who argues that the passions are "feelings that are accompanied by pleasure or pain."(7) Following Aristotle, most accounts divide the passions in reference to good or bad objects and experiences. Good experiences, such as dining, are pleasing, whereas bad experiences, such as illness, are displeasing. Consequently, passions such as joy and grief are mirror images according to the good (pleasing) or bad (displeasing) nature of the experience.
A second feature of most theories is that there is a small foundational class of passions, which give rise to others through variation. The most popular foundational set includes four: joy, sorrow, hope, and fear. Although the earliest reference to these four is by Plato (Laches, 191C), Zeno of Citium and the Stoics are attributed with developing a theory of the passions based on these. The fullest surviving account of the Stoic theory is given by Cicero in his Tusculan Disputations.(8) In Cicero's discussion, the four most primitive passions are joy (laetitia), grief (aegritudo), desire (cupiditas or libido), and fear (metus). Joy and desire are the result of good circumstances whereas grief and fear are the result of evil. Further, joy and grief result when a circumstance or object is actually present, in contrast to desire and fear which arise when one anticipates a future circumstance. The division is as follows:
Present joy grief
Anticipated desire fear
In Roman and medieval philosophy, the Stoic division of the passions
was widely endorsed by writers such as Virgil, Augustine, and Boethius.(9)
With Thomas Aquinas, though, an alternative classification arose. Thomas
recognized that in one sense the four Stoic passions were the most primary
(insofar as present and future goods and evils culminate in them).(10)
However, Thomas saw a more basic division based on concupiscible and irascible
appetites of the soul.(11) Concupiscible
passions are aimed at some enjoyable object which is relatively easy to
obtain. Irascible passions, on the other hand, are aimed at more difficulty
obtained objects which involve overcoming obstacles. For Thomas, there
are eleven primitive passions which are classified as follows:
Concupiscible
love
hatred
desire
aversion
joy
grief
Irascible
hope
despair
courage
fear
anger
During the Renaissance and Enlightenment, treatises on the passions appeared regularly. In varying degrees, most followed the taxonomy of either the Stoic catalog, emphasizing present versus anticipated goods, or the Thomistic catalog, emphasizing the divisions of the soul. Anthony Levi argues that many 17th century French theories of the passions only superficially followed the classical theories. According to Levi, "Plato, Aristotle, Zeno, Pyrrho, Augustine and Thomas were frequently quoted, but they were not always adequately understood.... It was of greater importance for the theory of the passions to know, for instance, that Plato was considered to have put the 'irascible faculty' into the heart than to have an accurate exegesis of the Timaeus."(12)
H.M. Gardiner argues that Juan Luis Vives's De anima et vita (1555) was a pivotal work of the period, which represented a general rejection of the Thomistic catalog.(13) This is not entirely correct. Nicholas Coeffeteau's Table of Humane Passions (1621), for example, is based on Thomas's catalog. Also, Thomas Wright's The Passions of the Mind in General, (1601), reduces the primary passions to Thomas' six concupiscible ones.(14) Hobbes seems to follow Wright by offering the same six, plus appetite (Leviathan, Ch. 6). Even Locke in his short discussion of the passions (Bk. II, Ch. XX) owes some allegiance to Thomas's theory. Although Locke does not mention the concupiscible and irascible divisions by name, in paragraphs 4-8 he lists and discusses the concupiscible passions and in paragraphs 9-14 he discusses the irascible ones.(15)
From the other camp, several Renaissance authors held firmly to the four Stoic passions, such as Timothy Bright, Treatise of Melancholy (1586), La Primaudaye, The French Academy (English, 1586), John Davies, Nosce Teipsum (1599), John Davies of Hereford, Microcosmos (1603). It has also been argued that the Stoic theory was the dominant model of the passions in Renaissance logic, rhetoric, and literature.(16) Others writers, however, took liberties with the Stoic catalog. Descartes, for example, lists six primitive passions (joy, grief, love, hate, desire, wonder),(17) and Spinoza offers three (pleasure, pain, and desire).(18) Malebranche's catalog is particularly illustrative of these revised Stoic accounts.(19) Malebranche lists six primary passions, according to whether objects are present, anticipated, or unobtainable:
Present joy love sorrowful aversion
Anticipated desirous love desirous aversion
Unobtainable sorrowful love joyful aversion
As late as the mid 19th century, the Stoic theory of emotions was adopted by some with little alteration. Frederick A. Rauch in his Psychology (1841), for example, lists pleasure, pain, hope, and fear as the foundational emotions from which compound ones arise.(20) The end of the traditional catalogs of the passions was signaled by William James in his Principles of Psychology (1890). James argues that,
...the merely descriptive literature of the emotions is one of the most tedious parts of psychology. And not only is it tedious, but you feel that its subdivisions are to a great extent either fictitious or unimportant, and that its pretences to accuracy are a sham.
Instead of focusing on emotions as particular types of feelings which lend themselves to cataloging, James shifted emphasis to psychological mechanisms which are responsible for emotions in general.
Beginning with Hippocrates,(21) attempts had been made to explain the passions in terms of physiological phenomenon, such as the humors, vital spirits, vapors, temperatures, and pneuma. Descartes, for example, sought to explain the passions in reference to the production of blood and spirits by the bodily organs and the effects these spirits have on the brain. Similarly, Malebranche's explanation was based on the "degree to which the animal spirits are agitated, in their quantity and consistency, and their relation to the brain's fibers."(22) Hutcheson's response to this approach was to "Let Physicians or Anatomists explain the several Motions in the Fluids or Solids of the body, which accompany any Passion..." (Essay, p. 57). Hume echoes Hutcheson by reserving such accounts for "the sciences of anatomy and natural philosophy" (T 276).(23)
For Hume, passions are a type of mental perception. He distinguishes between perceptions in a genus and species fashion as follows:(24)
IDEAS
PERCEPTIONS <
OF SENSATION
IMPRESSIONS <
OF REFLECTION
Basic to this division is that ideas differ from impressions by being less lively copies of impressions. Further, impressions, what we broadly call "feelings," are either of sensation or of reflection. Those of sensation involve the five senses as well as bodily appetites such as hunger.(25) Impressions of reflection, by contrast, involve the passions.
The term "reflection" (in contrast to "sensation") can be found in Locke
who used it to mean introspection, or the ideas of the mind when it takes
notice of its own operations. For Locke, though, this does not involve
passions, which are for him "internal sensations." The word "reflection"
becomes associated with the passions in Hutcheson's Essay on the nature
and conduct of the passions and affections (1728). At the opening of
section three, Hutcheson adopts the Stoic catalog of passions as given
by Cicero, and explains how joy and sorrow arises:
For Hutcheson, reflecting on a sensation (such as the visual sensation produced by a new suit of clothes) is only the occasion for the passion of joy to arise. This differs from Hume's meaning, which is that reflective impressions follow from impressions of sensations either immediately, or "by the interposition of its idea" (T 276). Unlike Hutcheson, reflections for Hume are not simply the occasion for the passions to arise. Instead, the reflection is the passion.(26)
Hume's most complete categorization of the passions is in the opening
paragraphs of Book II. He begins by dividing impressions of reflection
(hereafter R-impressions) into categories of calm and violent. Beginning
with Aristotle,(27) it was noted that passions
could be calm or violent. Often, though, this meant that a particular
passion of anger, for example, could be calm on one occasion, or violent
on another. This is the understanding of calmness and violence found, for
example, in Edward Reynolds' A Treatise of the Passions and Faculties
of the Soule of Man (1640):
But British writers, such as Tucker, Hutcheson, and Price, have argued further for a class distinction between calm and violent passions.(29) For them, the term "affection" designated calm emotions which involve a rational perception of one's situation, and the term "passion" (taken in a narrow sense) designated violent emotions, often grounded in instincts. Hutcheson, for example, argues that the four Stoic primitives (desire, aversion, joy and sorrow) are pure affections, since "the purest Spirit, were it subject to any Evil, might be capable of them." Thus, these "arise necessarily from a rational Apprehension of Good or Evil." By contrast, passions such as fear and anger arise when "more violent confused Sensations arise with the Affection, and are attended with, or prolonged by bodily Motions" (Essay, p. 61-63). In short, for Hutcheson, the affections are calm since they are the products of an immaterial soul, and passions are violent since they arise from disturbances within a material body.
Hume also acknowledges a class distinction between calm and violent
R-impressions:
Although Hume follows Hutcheson by making "calmness" and "violence" function as class names, he also acknowledges with Reynolds that any R-impression token might be experienced either calmly or violently.(30)
To avoid confusion, I will use the term "generally calm R-impressions" in reference to Hume's class of passion types whose tokens are experienced calmly on most occasions. Similarly, the term "generally violent R-impression" will refer to the class of passion types whose tokens are experienced violently on most occasions. The benefit of this terminology is that the members of these classes are passion types and not tokens, hence membership will remain constant despite any variantly experienced tokens. From the above quote, generally calm R-impressions for Hume include feelings of beauty and deformity, and generally violent ones include love hate, grief, joy, pride and humility.
The second set of divisions at the beginning of Book II are those between
the direct and indirect R-impressions. Although in Book II Hume begins
his discussion with the indirect R-impressions, the direct ones are more
fundamental. His account of the direct R-impressions, which appears at
the close of Book II, is as follows:
When either good or evil is uncertain, it gives rise to fear or hope,
according to the degrees of uncertainty on the one side or the other. (T
439)
Good Object Evil Object
Present joy sorrow
Anticipated
hope
fear
Thus far, much of Hume's account of the passions is traditional. He adopts the traditional view that passions are polarized according to their association with good or evil. He is also following tradition by grounding all passions in a class of primitives which, for him, total eight: four direct (Stoic) passions (joy and grief, desire and aversion) and four indirect passions (pride and humility, love and hate). Much of Book II and the "Dissertation" is occupied with showing exactly how the other passions arise from these eight primitives.(31) These factors may have given the reviewers from the Literary Magazine and the Critical Review a first impression that Hume was merely following tradition throughout his account, and was therefore unoriginal.
But, Hume's debt to tradition ends with his explanation of the indirect
R-impressions. These passions, which include pride, humility, love and
hatred, are called "indirect" since they arise indirectly from feelings
of pleasure and pain. The psychological mechanism by which they indirectly
arise is what Hume calls a "double relation" of ideas and impressions.
That is, indirect passions arise only when there is an association of one
idea with another idea, accompanied by an association of one impression
(or feeling) with another impression. The feeling of "pride," for example,
is explained as follows:
Hume's accounts of other indirect passions, such as humility, love and hate, are variations on this theme. For example, humility differs from pride only in that the initial idea (or subject) produces pain instead of pleasure, (as would be the case if Smith's suit of clothes were poorly fitting). With love, the difference is that the object of love is an idea of someone else, as opposed to oneself.
A final group of R-impressions in Hume's account of the passions involves a distinction between instinctive and acquired passions. Throughout his writings, Hume refers to R-impressions which inexplicably arise from natural instincts. These involve desires relating to bodily appetites such as hunger, thirst, lust, and mental appetites such as benevolence, resentment, love of life, kindness to children, vengeance, happiness of friends, attachment to offspring and ambition (T 8, 217, 218, 417, 439, E 201, 301). Commentators have referred to these classes as primary (as opposed to secondary) R-impressions. Hume himself, though, does not use these terms.(32) These instinctive R-impressions are called "primary" since they are foundational passions (in the sense that they do not arise from pleasure or pain). Thus, Kemp Smith, who first suggested this division, defines primary R-impressions as "sheerly instinctive passions, arising from a natural impulse or instinct not founded on precedent perceptions of pleasure and pain." Secondary R-impressions, by contrast, require a prior pleasure or pain and are not foundational. These include more complex feelings such as love, hate, pride, and humility). Again, a primary/secondary distinction among passions is not unique to Hume. Hutcheson, for example, argued that natural propensities are responsible for passions such as fame, revenge, anger, and pity (Essay, pp. 61-65).
As shown above, R-impressions, for Hume, fall into categories of calm, violent, direct, indirect, primary, and secondary. Exactly how these divisions are related is not clear from Hume's account, and there is considerable disagreement among commentators about the exact divisions. The first organized outline of the R-impressions was suggested by Kemp Smith:(33)
PRIMARY
R-IMPRESSIONS <
INDIRECT
SECONDARY <
CALM
DIRECT <
VIOLENT
Kemp Smith's interpretation fails, though, since it runs counter to passages which imply that generally violent R-impressions may be indirect, as well as direct. For, under the heading of generally violent passions, Hume lists love, hatred, pride, and humility, which he later tells us are also indirect R-impressions (T 476). Kemp Smith, as seen above, gives no subdivisions of the indirect passions which might include generally violent ones.
To cover this problem, Ardal revised Kemp Smith's outline by adding calm and violent divisions wherever they were absent (Ardal, 1966, 10-11):(34)
CALM
PRIMARY <
VIOLENT
R-IMPRESSIONS <
CALM
INDIRECT <
VIOLENT
SECONDARY <
CALM
DIRECT <
VIOLENT
Ardal's outline has been adopted by several commentators, including Philip Mercer, Jerome Neu, and Terence Penelhum. In spite of its acceptance, this outline has problems. First, as Ardal admits, it is more removed from Hume's outline of R-impressions as presented in the first four paragraphs of Book II.(35) More importantly, though, Ardal ignores Hume's categorization of emotion types into groups of generally calm and generally violent, as discussed above.(36) The historical precedent in conjunction with Hume's own discussion indicate the importance of calm and violent as class distinctions.(37)
As a reaction against Ardal's departure from the text, Louis Loeb has most recently proposed an outline which more closely follows Hume's intentions in the opening of Book II:(38)
CALM
R-IMPRESSIONS <
DIRECT
VIOLENT <
INDIRECT
Although Loeb's account correctly divides the class of generally violent passions between the direct and indirect, his interpretation is not complete. First, Loeb fails to account for the primary and secondary R-impressions. This distinction is prominent in Hume's account of the direct passions:
Beside good and evil, or in other words, pain and pleasure, the direct passions frequently arise from a natural impulse or instinct, which is perfectly unaccountable. Of this kind is the desire of punishment to our enemies, and of happiness to our friends, hunger, lust and a few other bodily appetites. (T 439)
Here Hume argues that direct R-impressions arise in two manners: (1) from the pleasure of some good object, and (2) from some natural instincts.(39) The first group of direct R-impressions is secondary and involves the traditional explanation of the four Stoic passions. The second group clearly involve the primary passions. Thus, direct R-impressions must be divided between the primary and the secondary.(40)
A second aspect of Loeb's discussion which is inadequate is his account of the generally calm R-impressions. These he identifies as moral sentiments of pleasure and pain exclusively. But in the opening of Book II, Hume describes these as "the sense of beauty and deformity in action, composition, and external objects" with specific reference to "the raptures of poetry and music" (T 276). From this description, one can gather that calm R-impressions involve aesthetic emotions and value judgments. From his various discussions of aesthetic emotions in Book II, it is also clear that "the sense of beauty and deformity" entails other pleasures and pains. Some pleasures Hume mentions are those produced from the beauty of a fine suit of clothes (T 439), from praise (T 324), from the pursuit of truth (T 448), or from music (T 472). Contrary to Loeb, the complete list of generally calm pleasures and pains is endless as may be seen from the variety of pleasures which cause the direct and indirect R-impressions (see, for example, the causes of pride and humility, T 279, 288).
Further, these pleasures and pains exhaust the category of generally calm R-impressions. For, all of the R-impressions discussed in Book II, save R-pleasures and pains, are either direct or indirect. And these are generally violent. This leaves these pleasures and pains as the only remaining candidate for the generally calm R-impressions. Although all generally calm R-impressions are pleasures and pains, not every pleasure and pain is a generally calm R-impression. Hume also speaks of some pleasures and pains which are impressions of sensation (or S-pleasures and pains; see T 192, 275, DP 139). S-pleasures and pains arise from the application of objects to our bodies, such as a knife cutting one's skin.(41)
The division of R-impressions suggested by the above discussion is as
follows:
Generally Calm (all and only R-pleasures and pains)
R-Impressions <
Primary (from instinct)
Direct <
Secondary (four Stoic passions)
Generally Violent <
Indirect (love, hate, pride, humility)
So far, I have argued that key aspects of Hume's classification of the passions follow tradition. I have also offered an account of his classification which addresses inadequacies of previous theories. In addition to disputes about Hume's classification of the passions in general, there is a related dispute about Hume's use of the word "passion." Kemp Smith holds that Hume uses "passion" broadly, but not in reference to pleasure and pain (Kemp Smith 162). Ardal's view is that Hume's use of "passion" is so broad that it includes all R-impressions (Ardal, 1966, 8-11). Loeb argues that, properly speaking, "passion" refers to only the violent R-impressions (Loeb 396). But Hume's meaning is mixed. In some contexts he uses "passion" to include both calm and violent R-impressions (T 419, 422, 437, 583). Yet, following others, such as Tucker, Price, and Hutcheson, he has a more narrow meaning, referring only to violent R-impressions (T 276, 437).(42)
In such contexts, "passion" colloquially refers to the violent R-impressions in a manner similar to the way "reason," for Hume, colloquially refers to the calm R-impression (T 419, 437, 583, cf. 470).
A second and related problem involves Hume's use of the phrase "impression of reflection." Are R-impressions and passions (in the broad sense) synonyms for Hume? Problems of classification occur with both the will and the feeling of causal anticipation involved in the idea necessary connection. Regarding the will, Hume argues that the will is not, properly speaking, included among the passions, yet is an "internal impression we feel and are conscious of, when we knowingly give rise to any new motion of our body, or new perception of our mind" (T 399). Regarding the feeling of causal anticipation, in the Enquiry, Hume refers to "the sentiment or impression from which we form the idea of power or necessary connexion" (E 74). He would also seem to classify this feeling as an "internal impression" (T 64). The Treatise presents further difficulties by describing the feeling as "some internal impression, or impression of reflection" (T 165). Unfortunately, the context of this passage does not make it clear whether internal impressions are identical to R-impressions.
Two interpretations emerge: (1) internal impressions, (which include the will and expectation), are distinct from R-impressions (which are the passions), and, (2) R-impressions are identical to internal impressions and include the passions, as well as the will and the feeling of causal anticipation. The first of these interpretations seems preferable. First, at T 8, Hume describes R-impressions as "passions, desires, and emotions," none of which seem proper classifications for the will and the feeling of causal anticipation. Second, Hume's use of the term "internal impression" is restricted to discussions of the will and causal anticipation, and never used in reference to the passions. Finally, Hume's catalog of R-impressions at T 276, and his subsequent analysis of them in Book II, do not classify the will or causal anticipation under the two key divisions of the R-impressions (i.e. the generally calm and generally violent).
1. References to Hume are from A Treatise of Human Nature, (Oxford University Press, 1978), hereafter designated as T, Enquiries Concerning Human Understanding and Concerning the Principles of Morals, (Oxford University Press, 1975), ed. L.A. Selby-Bigge and P.H. Nidditch, hereafter designated as E, and "Dissertation on the Passions," from David Hume: The Philosophical Works, (London: Longmans, 1886), ed. T.H. Green and T.H. Grose, hereafter designated as DP. Although the Treatise's account of the passions is consulted by most Hume commentators today, Hume's contemporaries were familiar almost exclusively with his "Dissertation on the Passions."
2. The Literary Magazine: or, Universal Review, 1757, Vol. 2, pp. 32-36.
3. Critical Review, 1757, Vol. 3, pp. 97-107, 209-216.
4. In Book II of the Treatise Hume comments that "My hypothesis is so simple, and supposes so little reflexion and judgment, that 'tis applicable to every sensible creature; which must not only be allow'd to be a convincing proof of its veracity, but, I am confident, will be sound an objection to every other system" (T 328).
5. William Rose, Monthly Review, 1757, Vol. 16, pp. 122-139.
6. The term "passion" connotes that it is passively affected by external circumstances. The derivation of "passion" from the Latin "passio" (being acted upon) is thus appropriate to its meaning.
7. Aristotle, Ethics (1105b20). Writers on the passions do not usually list pleasure and pain as passions, but, instead as feelings in a more general sense.
8. Tusculan Disputations (Cambridge: Loeb Classical Library, 1949), Book IV, 6, 11
9. Virgil lists joy, grief, fear, and desire. Aeneid, Bk. VI, vs. 733. Citing both Virgil and Cicero, Augustine lists the four most common emotions as joy (laetitia), grief (tristitia), desire (cupiditas), and fear (timor) The City of God, Bk. XIV, Ch. 3 and 5. Boethius lists joy (gaudium), grief (dolor), hope (spes), and fear (timor) De Consolatione, I, 7.
10. Thomas Aquinas, Summa Theologica, 1a2ae, Q. 25, 4.
11. Like the four Stoic passions, the irascible and concupiscible distinction also traces back to Plato. In the Timaeus (69), Plato argues that the gods fashioned our soul with the passions distributed in two places. At our midriff are passions which are "terrible and irresistible" (i.e. the irascible). These include pleasure, pain, rashness, fear, anger and hope, and all-daring love. At our chest were placed passions which include "courage and passion" as this part of the soul "loves contention" (i.e. the concupiscible).
12. Anthony Levi, French Moralists (Oxford: Clarendon Press, 1964), p. 7.
13. H.M. Gardiner's Feeling and Emotion, a History of Theories, (New York: 1937), pp. 128-131.
14. Wright also follows Stoic tradition by distinguishing between his six passions according to whether they are present or anticipated.
15. John Locke, Essay Concerning Human Understanding, (Oxford University Press, 1975), ed. P.H. Nidditch, pp. 229-233.
16. Rolf Soellner, "The Four Primary Passions: A Renaissance Theory Reflected in the Works of Shakespere," Studies in Philology, 1958, Vol. LV, pp. 549-567.
17. Rene Descartes, The Philosophical works of Descartes, (Cambridge: Cambridge University press, 1977), p. 362.
18. Benedict De Spinoza, The Ethics (New York: Dover, 1955), p. 175.
19. Nicolas Malebranche, The Search After Truth (Columbus: Ohio State University Press, 1980), p. 375.
20. Frederick A. Rauch, Psychology, (New York: M.W. Dodd, 1841). With surprisingly little change from Cicero's discussion, Rauch notes how joy differs from mere pleasure: "The latter has only reference to the present, joy always more or less to the future, and hope is one of its necessary constituents."
21. See Gardiner's account of Hippocrates and his influence Feeling and Emotion, pp. 4-7, 86-88.
22. Malebranche, The Search After Truth, p. 374.
23. Despite this reluctance, though, Hume implies that there is a biological account of the principles of association and thereby a biological account of the passions (cf. T 60).
24. Although recognizing Hume's basic divisions of perceptions, R.W. Altman argues that, as a matter of philosophical method, Hume "first makes broad distinctions and then colors in the various shades of meaning or subtle differences between terms or notions." In one particular passage Altman finds Hume "minimizing the distinction between impressions and ideas almost to the point of identifying them," "Hume on Sympathy," Southern Journal of Philosophy, (1980), pp. 124-125.
25. Hume says little about the nature of impressions of sensation, or S-impressions, reserving this for the study of anatomists and natural philosophers (T 8). But he does tell us at T 192 that there are three kinds of S-impressions: (1) impressions relating to figure, bulk, motion, and solidity of bodies (similar to the notion of primary quality), (2) impressions relating to color, taste, sound, heat and cold (similar to the notion of secondary quality), and (3) pleasures and pains that arise from the application of objects to our bodies (which he also refers to as "bodily pleasures and pains" T 275; see also DP 139).
26. More precisely, reflective impressions are passional reflections of the sensation itself, or of the ideas of these sensations.
27. Aristotle, De Anima (403a20).
28. Edward Reynolds, A Treatise of the Passions and Faculties of the Soule of Man, (London, Robert Bostock, 1640), p. 60. Descartes also distinguishes between the lesser and more violent passions.
29. Abraham Tucker, The light of nature pursued, Pt. I, Ch. XXI, Sec. I, Richard Price, Review of the principal questions and difficulties in morals (1758), pp. 117.
30. Granting a class distinction between the calm and violent R-impressions, Hume's explanation of the causes of the violent passions in Book II (T 418-438) must be an account of the vivacity of token passions, and not of passion types.
31. For the indirect passions, Hume argues that esteem and contempt arise from love and hate (respectively) plus sympathy (T 357, 362); benevolence, anger, pity, malice and envy arise from love or hate plus desire or aversion (T 367, 372; pity and malice are also said to "imitate" love and hate); respect and contempt are noted to be a balance between pride and humility, and love and hate. Hume also notes that ambition (and perhaps vanity by implication cf. 276) has the same explanation as pride, only with a more limited object (T 300). Lastly, he notes that tenderness, friendship, intimacy, esteem and goodwill are love only of a different shape (T 448). Regarding the direct passions, Hume argues that hope and fear are each varying mixtures of grief and joy. Further, terror, consternation, astonishment, anxiety and "other passions of that kind" (perhaps despair and security by implication, T 276) are said to be species and degrees of fear (T 447).
32. See Norman Kemp Smith, The Philosophy of David Hume, (London: Macmillan and Co, 1941), p. 168; Alfred Glathe, Hume's theory of the Passions and of Morals, (Los Angeles: University of California Press, 1950), p. 30; Pall S. Ardal, Passion and Value in Hume's "Treatise", (Edinburgh: University Press, 1966), p. 10; and Philip Mercer, Sympathy and Ethics, (Oxford: Clarendon Press, 1972), p. 22. To avoid confusion, the primary/secondary distinction discussed here is not the same distinction as that which Hume makes between "original" and "secondary" impressions at T 275; as the context indicates, the latter clearly mirrors the distinction between sensation and reflection.
33. Kemp Smith, p. 168. Glathe adopts Kemp Smith's outline of the passions; Glathe, p. 30.
34. Mercer, Jerome Neu, and Terence Penelhum all follow Ardal's outline of the passions. Mercer, pp. 22-23; Neu, Emotion, Thought and Therapy, (London: Routledge and Kegan Paul, 1977), p. 68; Penelhum, Hume, (New York: St. Martin's Press, 1975) p. 93, note 4.
35. Ardal writes, "I want to make it quite clear that the classification of the passions that I gave in Passion and Value was meant to be an improvement upon the account Hume himself gives at the beginning of Book II. I therefore completely agree... that the reading of the opening of the pages of Book II of the Treatise would suggest a classification of impressions of reflection different from the one I suggested." Pall S. Ardal, "Another Look at Hume's Account of Moral Evaluation," Journal of the History of Philosophy, (1977), p. 410.
36. Ardal does makes a theoretical distinction between "calm" and "violent" as class names versus tokens (Passion and Value, p. 104). However, the calm-violent distinction in his outline and discussions reflect calm versus violent tokens.
37. At T 417, Hume refers to direct desires (both primary and secondary) which are calm. Ardal and Penelhum take this to imply that calm passions include some direct passions (Ardal, 1966, p. 97; Penelhum, p. 93). A better interpretation would be to understand these calm desires as tokens, and not as types. This coincides with Hume's discussion further down that same page where he notes that there are violent desires of the same kind as these calm desires (b 417-t 418).
38. Loeb's argument is that, first, the division of mental events in the beginning of Book II consistently assigns the term "passion" to only violent R-impressions. Second, the most straightforward reading of paragraphs three and four makes directness and indirectness an exclusive distinction of the passions, and, hence, of violent R-impressions. Lastly, since these paragraphs contain the only exhaustive division of mental events found in the Treatise, the division "should be taken as canonical." Louis Loeb, "Hume's Moral Sentiments and the Structure of the Treatise," Journal of the History of Philosophy, (1977), p. 395.
39. This distinction between primary and secondary direct R-impressions is also found at T 417-418.
40. It is important to note that the primary/secondary distinction is not particularly relevant to indirect R-impressions. For, technically, all indirect R-impressions are secondary because they must arise from pleasures and pains (see Hume's second experiment, T 334). It is, therefore, trivial to think of indirect R-impressions as secondary (as opposed to primary).
41. Stephen Hudson argues that two features distinguish R (or reflective) pleasures from S-pleasures: (1) R-pleasures cannot occur without one paying heed to, or being aware of some object (for instance, a suit of fine clothes), and, (2) R-pleasures are not localizable in the manner that S-pleasures are. "Humean Pleasures Reconsidered," Canadian Journal of Philosophy, (1975), 548-549.
42. Eugene Miller notes that Hume's use of "passion" in his essay "Of the Delicacy of taste and Passion" is narrow, referring only to violent R-impressions. Miller implies, however, that Hume's normal use of "passion" is broader, including calm R-impressions as well. Essays: Moral, Political and Literary, Ed. Eugene Miller, (Indianapolis: Liberty Classics, 1985), 3, note.