// While an emotion is generally described as a physiological response to an external stimulus (e.g. fear in the face of danger), a feeling is more generally described as a state of consciousness, or an internal, more subjective representation of emotion (http://en.wikipedia.org/wiki/Emotion) - for the purpose of discussion, I will treat them both as being equivalent, in the sense that both are subjective experiences. //
Our individual perception of feelings and emotions towards an object are always true. That is to say, If I am able to perceive my feelings regarding apples, I can know the truth value of the proposition "I like apples", since whatever definition I use for "like" is adequate in the personal realm, whether I consider the proposition to be true or false. What one means by "like", "love" or even "hate" is different from the emotion, or the experience, itself. In fact, at this point, it would be confusing to state that we experience "different emotions", but rather that we experience "emotion" - this definition encompasses all the complex states that an individual might experience, disregarding any labels usually given by us. Although our experience of emotion is, in this sense, absolute and personal, the definitions that we use to describe emotion are not absolute - this raises a problem of subjectivity in describing and communicating our personal experiences to others, and I will try to clarify this issue with an example. If an individual X tells individual Y that he feels love towards individual Z, but Y notices that X is generally aggressive, spiteful, and murderous when he is with individual Z, Y will have a hard time believing that X really loves this individual Z. If the feeling/emotion "love" exists, and X has no doubt that he loves Z, why does Y doubt? That is because what we humans conveniently labelled as "love" is nothing more than an expression of feeling in the form of actions. Even though we perceive and experience emotion, potentially in an unique way which cannot be replicated in other individuals nor completely evaluated, our actions are a function of our subjective experiences - if there is a pattern in the actions taken by a high number of individuals when they experience a certain biological state, then it is convenient to label this biological state. Y will probably not believe that X loves Z because the actions taken by X when he experiences Z do not fall within the set of actions caused by what he previously labeled as "love". To say that a mother loves her child is to say that mothers will generally take certain actions, similar for different mothers, towards each of their child that we therefore label as "love". Even though we cannot access individually the experience of each of these mothers and conclude that these are exactly the same, there is a certain pattern of actions taken on behalf of these emotions that is similar. If a mother states that she loves her child, but abuses, beats down, and violates its own child, most people will not believe her claim - because what we label as "love" is the cause of radically different actions than those taken by this same mother. We are thus at a crossroads: on the one hand, we have the feeling or emotion perceived and experienced by the individual, which is potentially unique; on the other hand, we have the actions taken on behalf of this emotion, which are recognized as a pattern by humans and therefore labeled.
If we wish to systematize and properly define a certain emotion, we are stuck between individuality and actions taken and recognized by others on behalf of emotional states.
Should we walk the long, treacherous path of individual definitions of emotion, or should we simply label it as a pattern, or set, of actions generally observed in individuals? If the former, we probably won't find a satisfying definition that is in accordance with every person and accurately depicts each and every wave in an ocean of sentiments; if the latter, another problem is raised - what if we take no noticeable actions on behalf of the emotions that we feel so that these cannot be evaluated by anyone except ourselves? And if we were mere spectators, how would we be able to distinguish such an individual from an emotionless machine?
I have thought about some possible definitions of emotion that attempt to alleviate some of these problems, but at this time I won't dwell on it. It needs more thought and clarification before it can be presented as something worth considering. Moreover, the future development of science, particularly of the fields of neurology and psychology, might help us establish a link between the personal and the external realm.