There is a wide agreement among philosophers, both contemporary and ancient, that emotions are capable of being appropriate and inappropriate or having and lacking warrant. An emotion of fear, for instance, is warranted only if its object is really dangerous for the emoting subject, or if the subject has good reasons to view the object as dangerous. Most contemporary philosophers refer to this kind of warrant by the notion of appropriateness or fittingness that is often taken to be an analogue of truth in the emotional domain. However, if we allow an analogue between appropriateness and truth, why not go all the way and argue that emotions are capable of being true and false? Many theorists appear to flirt with this idea when they suggest that appropriate emotions "enable us to get things right" (Goldie, 2004, p. 99) by "properly tracking those properties [funny, shameful, etc.] of which they purport to be perceptions." (D'Arms & Jacobson, 2000a, p. 69) After all, such locutions as 'getting things right' or 'getting the world right' and 'tracking' are usually applied to such discourses where a truth predicate is available. Yet only a few philosophers, most notably Ronald de Sousa (2002, 2004, 2011), have risen to the challenge to offer a plausible account of emotional truth.
There are several reasons for why philosophers have avoided the notion of truth in the emotional domain. One concerns the nature of emotions. Humean philosophical psychology attributes truth value to sentences and propositional attitudes with assertoric content and the mind-to-world direction of fit, such as beliefs, thoughts, and judgments. Therefore, it seems that if emotions are truth-apt, they must be reducible to propositional attitudes. Indeed, hope and fear often figure as examples of propositional attitudes, but such is not the case with many other emotions, including love and hate.
Another prominent worry focuses on the notion of truth. Truth and objectivity are so closely related that it is difficult to see how a domain whose standards of epistemic warrant defy objectivity could qualify as truth-apt. No doubt, evidentially unconstrained truth is beyond the reach of emotions; it would be absurd to suggest that things could, for instance, be funny without an amused human response to them, whether actual or idealized, as Crispin Wright (1992) points out. However, even an anti-realist account of truth in terms of warranted acceptability of some kind seems unattainable in the emotional domain, because it is hard to make sense of warranted acceptability without invoking irreducibly fragmented communal or personal standards of appropriateness. Therefore, it appears counterintuitive to look for 'objective' affective properties of objects and events.
In this chapter, I attempt to respond to both main concerns about the possibility of emotional truth. Thus I will argue that on an adequate understanding of the nature of emotions, they are truth-apt mental states, even if they cannot be reduced to propositional attitudes. I will also suggest that an anti-realist theory of truth provides a promising model for emotional truth. I start by analysing de Sousa's view of the nature and truth of emotions which I find promising yet insufficient and sketchy as it stands. The ensuing discussion focuses on the problems of de Sousa's view together with my elaborations of it.
-  Nussbaum (2001, p. 46) also suggests that emotions are evaluative judgments with the mind-to-world direction of fit and therefore, "like other beliefs, [they] can be true or false" However, she doesn't elaborate this point further.