Skip to main content
SpringerLink
Log in

Moral Sensitivity and Desire Attachment: In What Sense are they Constituents of One’s Rational Profile?

  • Published:
Acta Analytica Aims and scope Submit manuscript

Abstract

A quantitative interpretation is given of the (in)coherence that moral agents experience as a tension between their ordered moral judgments over n physically incompatible actions, and the competitive ordering of motivating intensities (or, desires). Then a model describing one’s tendency to reduce the experienced in-coherence is constructed. In this model, moral sensitivity (S) and desire attachment (e) function as primitives that motivate from opposing perspectives the reduction of incoherence. Two distinct sub-processes of this reduction are therefore initiated by (S) and (e) co-ordinated (more or less efficiently) by the agent’s degree of rationality (R) characteristic of her capacity to handle such internal tensions. This process ends when a new equilibrium between what motivates and what resists (further) reduction has been reached. A macro-equilibrium is described involving (R) constrained by weakness-of-will (W w ). A reinterpretation of the Aristotelian characters (enkratês, akratês, etc.) and an exegesis of Hume’s ‘Calm Passions’ follow as applications.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Smith (2004)

  2. This is a simplified reference to the relevant text for I have omitted the complication of the ‘confidence to the means’ parameter that I believe does not affect the basic idea and structure of the described incoherence except perhaps on occasions to ‘explain it away’ as Smith claims it does (ibid, p.100).

  3. Edward Harcourt [2004] (see M. Smith (2004)), elaborates upon several counter examples on Smith’s thesis that the coherence as he describes it is a requirement of instrumental rationality. What is of direct relevance to the issue raised in this introduction is Harcourt’s emphasizing that irrationality cannot be consistently assigned to ‘transmissions’ of desires from ends to necessary means; coming to desire what on occasions seems impossible to do, for example, desiring to climb up a very thin branch to pick the last fig, is, Harcourt insists, something that one might, or, might not do (emphasis mine). One cannot be criticized for irrationality, he says, if one desires the impossible, as for example ‘climbing up the branch’, but nor can be criticized for letting one’s desires “...to wander pointlessly, down further pathways where (one) believes they will remain unfulfilled” (pp. 122, 123).

  4. Smith should be happy with this internalist confession!

  5. An order of moral preferences (or, judgments)<J1...Jn>, as interpreted here orders the ‘rightness’ of actions that are at time t physically incompatible (e.g., nursing one’s grandma, and going for vacation with the family at the very same period).

  6. In Koutoungos (2005) I have argued that failing to allow for a quantitative interpretation of moral coherence is a serious weakness of Smith’s ‘Weak Internalism’ vis-à-vis his attached explanatory claims of the so called ‘striking facts’, referring to “changes of one’s moral beliefs that are observed to lead in changes of one’s moral behaviour”, and that this weakness extends even to a failure of internalism to differentiate substantively from externalism. The quantitative interpretation of coherence (and of moral worth) I have provided in this former paper will here be reviewed as a result of the introduced dynamics though the overall philosophical justification of the quantitative approach is the same.

  7. See Appendix 2.

  8. On the contrary, any effort to affect the order of judgments ‘is measured’ only by the result of actual changes brought in that order, no estimate of this effort prior to the resulting changes would make any sense here.

  9. And the asymmetry remains even after the introduction of a possible quantification (analogous to intensities) in the orderings of judgments. The asymmetry remains because such intensities could be understood here only epistemically, namely, as subjective probabilities held by the agent that her judgment Jn, for example, has correctly been placed before (or, after) her judgment Jm. Of course this is quite plausible and we must confess that the phenomenon of obsolete revision, the revision of these subjective probabilities that fall short to bring about a change in the order, can be repeated. The observation is intriguing indeed and requires further attention. For the purposes of this paper, however, I believe we can safely count on the radical distinctness between the epistemic nature of subjective probabilities and the psychological nature of the intensities of desires; revising the order of moral judgments even through revising the subjective probabilities by which they are so ordered remains a process radically incommensurable to the revision of the intensities of the corresponding desires.

  10. In other words, accepting along with Hume that beliefs (as moral judgments are interpreted from a standard cognitivist point of view) and desires are ‘distinct existences’, we cannot entertain the possibility of a function that translates immediately revision results measured in the perspective of judgment disorder DJ into revision results in the perspective of the desire disorder Δm.

  11. cf. Smith (1994), pp. 159–160

  12. Although Smith never raised the issue explicitly, he could not have accepted the rationality of the judgment disorder adjustment for it would deprive him of the only visible advantage over externalism: whether, as he claims, his explanatory link (judgment → motivation) depends, unlike the externalist ‘psychological’ desire for moral rightness, on a rational desire (cf. Koutoungos (2005) pp. 72–3).

  13. Such issue of course never arises in connection to the expressivist interpretation of moral judgments since there is no ‘conceptual space’ for a distinction of judgments from their corresponding motivations. This distinction tends also to be trivialized in the case of moral subjectivism.

  14. For a detailed analysis of these ideas and of the fundamental heuristic (the mathematical description of which follows) in the area of belief revision resulting in elementary communication (beliefs considered as subjective probabilities assigned to propositions) see Koutoungos (2003).

  15. For reasons of simplicity the resisting factor R es is assumed constant.

  16. This (second-order) motivating attitude should be practically conceived as ‘weighting’ against the competition from the (first-order) motivating intensities (desires).

  17. Recall that the motivating desires that according to Smith ‘conceptually’ follow judgments are not to be understood as quantities. Yet, as I have already argued elsewhere, this is an inconsistency in Smith’s perspective for it is only when such motivations override competitors in terms of intensity that the link between changes in beliefs and changes in behavior becomes observable, allowing, as a result, reference to Smith’s ‘striking fact’.

  18. For the calculation of Δm see Appendix 3.

  19. d m , the ‘solution’ of the corresponding H1 equation, measures here the proportion of the accomplished disorder reduction – disorder perceived as Δm - when equilibrium has been reached between motivating (S, Δm) and resisting (e) factors. It can be easily observed that in the absence of resistance (e = 0) the accomplished disorder reduction goes all the way since d m  = Δm. Notice also that if the competing dispositions are equal (S = e), the disorder reduction rests half the way, and of course if no disorder is perceived (Δm = 0) no corresponding reduction takes place.

  20. It will turn out that the measure of the uniquely perceived incoherence in this setting cannot possibly exceed the maximum among the two possible one-sided ‘perceptions’ [max(Δm, DJ)] (see below Appendix 1).

  21. As it becomes clear from H2m and H2J, the proportions of reduced incoherence d m and d J can increase up to the point of becoming equal correspondingly to Δm and DJ, provided either e, or, S (but not both) are equal to zero.

  22. From H2m, d m  = Δm when e = 0. But then, as it follows from H2J, d J  = 0, and this is precisely what is consistent with the fact that after the complete elimination of Δm there is no DJ portion of incoherence hanging for further reduction (and reversely, complete elimination of DJ, leaves no portion of Δm hanging).

  23. Obviously this angle measuring the cooperation of the two trends cannot be conceived as independently observable; it is only a derived quantity reflecting the actual results of revisions that have reduced by so much the initial overall incoherence.

  24. Thanks to the structure of these H2 expressions the undesirable result maxdc = maxdm + maxdJ = Δm + DJ is avoided – obviously it makes no sense to add the two aspects of the basically unique incoherence.

  25. See Appendix 1.

  26. Such restriction would not have been possible had the resisting and motivating factors been structured instead as S+x, e+y.

  27. On the essential and technical differences of these deliberations recall section I.

  28. Yet, they do not necessitate the metric so far employed; any consistent metric surely suffices to reflect a single aspect’s ‘amount’ of incoherence as a proportion but it cannot force a comparison of the motivating impact of incoherence as ‘perceived’ in each of the two aspects with the other in terms of some particular metric. What does it really mean that Δm is smaller, or greater than DJ other than that the one has a greater motivating impact upon the agent (who experiences incoherence in this way) than the other?

  29. A secondary point, yet, one that might have kept worrying some readers all along, concerns the unit-system underlying the employed quantity-concepts. The answer to this very important technical point is here rather simple. In the structure of the fundamental heuristic applied throughout this exposition two kinds of quantities appear: the ‘distance’ (or, disorder) figuring as nominator, and the fraction multiplying this distance weight×(weight+resistance)]. Fortunately both these quantities (distance, and fraction) end up as unit-neutral: the introduced distances (as explained in the appropriate footnotes and elsewhere) are pure percentages, while the fraction contains always homogeneous underived quantities of the form a/(a+b) and so by definition is also unit-neutral. Further, a and b that interpret here motivation and resistance are homogeneous in the same sense that in physics force and friction are.

  30. Given the structure of the H2m and H2J, for S = 0 and e = 0 both dm and dJ become indeterminate; it follows from h that dc (as a function of dm and dJ), remains indeterminate, and that by this indeterminacy R is also infected given its (macro) dependence on dc (HR).

  31. For it must be clear that for any type of enkratês the elimination of the internal tension against contrary to judgment desires is a morally preferred course of action; it is therefore a sign of akratic behaviour if not taken. I believe that M. Smith would endorse this approach against the ‘static’ enkratês as strongly echoing his judgment internalism.

  32. A useful source of this and other relevant ideas by Hume is a paper by Jane L. Mcintyre, Mcintyre (2006).

  33. A rather important technical point must be posed and resolved now: in the process of incoherence reduction represented by the Δm path, in other words, by the path of revising the order of intensities we must allow in view of the preceding discussion of the calm passions that the ‘strength’ parameter, not only the ‘violence’ (as it is signified by the term ‘intensity’), is adequately taken into consideration. This can be accounted as follows: in the process of deliberation aiming towards Δm revision, the ‘strength’ interferes as a lower level resistance of its and every entry in the ordering of desires, the resistance against lowering its original position in the ordering.

  34. (as cited by Mcintyre) “Tis evident passions influence not the will in proportion to their violence, or the disorder they occasion in the temper; but on the contrary, that when a passion has once become a settled principle of action, and is the predominant inclination of the soul, it commonly produces no longer any sensible agitation. As repeated custom and its own force have made every thing yield to it, it directs the actions and conduct without that opposition and emotion, which so naturally attend every momentary gust of passion. We must, therefore, distinguish betwixt a calm and a weak passion; betwixt a violent and a strong one” (T 2.3.4..1; SBN 418–419).

  35. Daniel Shaw interprets calm passions as unactualized dispositions. Surely this is a sort of solution for the problem of introspective access one is expected to have for one’s own calm passions had the latter been of an emotional stuff no matter how faint – the problem for Hume’s empiricism would be in this case that calm passions are no more than a fiction of an ad hoc hypothesis. But then the new problem as Shaw himself admits is that it is hard to understand this unactualized disposition “acting as a causal factor at the time of action”. He proposes as a result to consider in addition some “occurrent basis”, “some brain state underlying it” filling the causal link (of the unactualized disposition) with action and he concludes regarding “...as a further empirical question whether or not such unactualized dispositions are ever operative in the absence of some underlying occurrent basis” (ibid, pp. 175–176). I believe that Shaw’s introduced ‘occurrent basis’ or ‘brain state’ might then just as well admit the interpretation proposed here as instances of emerging resistance (in conditions of internal competition of passions) that allows whatever preceding state of passion to be transformed into a causally efficacious one. The latter interpretation has the clear advantage that the additional condition required to establish the causal link is totally independent from any properties, or features of the calm passion (whether faint or unactualized) under consideration; such dependence would perpetuate the mystery of the interconnection of these properties, or features with the passion itself so that the actualization, or causal adequacy of it in given circumstances is explained.

  36. J. Mcintyre also emphasizes the significance of this distinction but she doesn’t think Hume help us to understand it: “Hume asserts the importance of distinguishing strong passions from violent ones, but he does little to explain how calm passions gain strength without becoming violent” (ibid, 399).

  37. These intuitions are to be found in the explanations of strength by sympathy. These explanations, as J. Mcintyre reports (ibid, p. 400), are presented also in a negative sense; sympathy is presented (in EPM 9.1.9; SBN 275) as partly explaining “popular tumults, seditions, factions and panics. Also, more significant intuitions concerning the motivating force of communication processes that end up increasing the strength of a calm passion are found in comments stressing the evaluation of the opinion of others (that we are, in other words) “..” (T 2.1.11.11; SBN 321), and also in the importance of sharing sentiments which, being similar reinforce each other (T2.1.11.6; SBN 318).

  38. See concluding remarks in section 4.

References

  • Koutoungos, A. (2003). The possibility of partial agreement. Philosophical Inquiry, 25(1–2), Winter-Spring, Thessaloniki.

  • Koutoungos, A. (2005). Moral coherence, moral worth, and explanations of moral motivation. Acta Analytica, 20, 3 summer.

    Article  Google Scholar 

  • Mcintyre, J. L. (2006). Strength of mind: Prospects and problems for a Humean account. Synthese, 152, 393–401. doi:10.1007/s11229–006–9005–4.

    Article  Google Scholar 

  • Smith, M. (1994). The moral problem. Blackwell: Oxford.

    Google Scholar 

  • Smith, M. (2004). Instrumental desires, instrumental rationality, (I), M. Smith, pp. 93–109, (II), E. Harcourt. pp. 11–129: Supplement to the Proceedings of the Aristotelian Society, (78).

  • Stroud, B. (1977). Hume (The arguments of the philosophers). Routledge.

  • Shaw, D. (1989). Hume’s theory of motivation. Hume Studies, 15(1), 163–183 April.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Humanities, School of Applied Mathematics and Physics, National Technical University of Athens, Zografou Campus, Zografou, Athens, Greece

    Aristophanes Koutoungos

Authors
  1. Aristophanes Koutoungos
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Aristophanes Koutoungos.

Appendices

Appendix 1

Prove that any fraction F with the structure (ax+by)/(x+y) takes its highest value from the highest among a, and b.

  1. 1.

    Let F > a:

    $$\begin{array}{*{20}l} {{{\text{F}} = {{\text{ax}}} \mathord{\left/ {\vphantom {{{\text{ax}}} {{\left( {{\text{x}} + {\text{y}}} \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}} + {{\text{by}}} \mathord{\left/ {\vphantom {{{\text{by}}} {{\left( {{\text{x}} + {\text{y}}} \right)} - {\text{a $>$ 0}} \Rightarrow }}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)} - {\text{a $>$ 0}} \Rightarrow }} \hfill} \\ {{{{\text{ax}}} \mathord{\left/ {\vphantom {{{\text{ax}}} {{\left( {{\text{x}} + {\text{y}}} \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}} - {\text{a}}{{\left( {{\text{x}} + {\text{y}}} \right)}} \mathord{\left/ {\vphantom {{{\left( {{\text{x}} + {\text{y}}} \right)}} {{\left( {{\text{x}} + {\text{y}}} \right)} + {{\text{by}}} \mathord{\left/ {\vphantom {{{\text{by}}} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow }}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow }}}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)} + {{\text{by}}} \mathord{\left/ {\vphantom {{{\text{by}}} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow }}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow }}} \hfill} \\ {{{\text{a}}{{\left( {{\text{x}} - {\text{x}} - {\text{y}}} \right)}} \mathord{\left/ {\vphantom {{{\left( {{\text{x}} - {\text{x}} - {\text{y}}} \right)}} {{\left( {{\text{x}} + {\text{y}}} \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}} + {{\text{by}}} \mathord{\left/ {\vphantom {{{\text{by}}} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow }}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow }} \hfill} \\ {{{{\text{ - ay}}} \mathord{\left/ {\vphantom {{{\text{ - ay}}} {{\left( {{\text{x}} + {\text{y}}} \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}} + {{\text{by}}} \mathord{\left/ {\vphantom {{{\text{by}}} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow {\text{b $>$ a}}}}} \right. \kern-\nulldelimiterspace} {{\left( {{\text{x}} + {\text{y}}} \right)}{\text{ $>$ 0}} \Rightarrow {\text{b $>$ a}}}} \hfill} \\ \end{array} $$
  2. 2.

    Let F > b

Following the same steps as above ⇒ a > b.

Therefore, F cannot possibly exceed both a and b.

It follows that its max. value=max (a,b).

Appendix 2

Let Δ measure the maximum ‘distance’ of two different orderings of n elements in terms of the amount of simple steps required to restore fully the identity of the two orderings. It follows that:

  • \(\Delta = {\text{n}} \times \left( {{\text{n}} + 1} \right) - 2 \times _1 \Sigma ^{\text{n}} \left( {\text{N}} \right)\), for all odd n, and

    \(\Delta = {\text{n}}^{\text{2}} - 2 \times _1 \Sigma ^{{\text{n}} - 1} \left( {\text{N}} \right)\), for all even n. N, in both cases is the set of odd positive integers.

Let now δ0 measure a simple distance (not the maximum) between two different orderings of n elements. A percentage of this distance (D0) is obtained when δ0 is divided by Δ (D0 = δ0/Δ). A more realistic difference (D), however, between two orderings is obtained if the distance of their upper parts (ui) is given some extra weight. This can be calculated again as a percentage claimed by ui/Δ from the available remainder distance 1-D0 (=Δ-δ0/Δ):

$${{\left[ {\delta ^0 + \left( {\Delta - \delta ^0 } \right) \times {\text{u}}^{\text{i}} \times \Delta ^{ - 1} } \right]} \mathord{\left/ {\vphantom {{\left[ {\delta ^0 + \left( {\Delta - \delta ^0 } \right) \times {\text{u}}^{\text{i}} \times \Delta ^{ - 1} } \right]} {\text{D}}}} \right. \kern-\nulldelimiterspace} {\text{D}}}.$$

[Notice that the maximum ‘weight’ added to the basic distance δ0 obviously corresponds to the maximum value of ui (ui = δ0), and that in this case the above D estimate=D0 × (2–D0) consistently remains ≤ 1 (D = 1 for maxδ0 = Δ)]

Appendix 3

Let the following five desires-motivations m1, m2, m3, m4, m5 that correspond to the ordered judgments J1, J2, J3, J4, J5 be ordered according to their own relative intensities, independently, that is, of the ordered moral preferences as in <J1, ..., 5>: m3 (=0.8), m2 (=0.6), m1 (=0.3), m4 (=0.2), m5 (=0). Δm can be now calculated as a percentage, the percentage of the volume of the disordered (relatively to judgments) intensities over the total volume of those intensities. m3, for example, holding the first place instead of the third, exceeds by 0.5 the intensity of m1 that holds the third place instead of the first, m2, still holds its proper (2nd) place, and again, m1 must also increase by 0.5 to be placed first (m4 and m5 are in order).

Therefore the ‘disordered volume’ (0.5+0.5 = 1) divided by the total volume=1.9, gives Δm=0.526. Of course more sophisticated algorithms taking into account the differentiation of the disorder between the upper and lower parts of the m ordering can be explored.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Koutoungos, A. Moral Sensitivity and Desire Attachment: In What Sense are they Constituents of One’s Rational Profile?. Acta Anal 23, 125–145 (2008). https://doi.org/10.1007/s12136-008-0025-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12136-008-0025-1

Keywords

Search

Navigation