My Aesthetica: 2011

"Every sort of senselessness or cowardice or dissipation or harshness that goes to excess is either animal-like or disease-like."
~Aristotle

For Aristotle, akrasia, "unrestraint", is distinct from animal-like behavior because it is specific to humans and involves conscious rational thinking about what to do, even though the conclusions of this thinking are not put into practice. When someone behaves in a purely animal-like way, then for better or worse they are not acting based upon any conscious choice.
Returning to the question of anger or spiritedness (thumos) then, Aristotle distinguishes it from desires because he says it listens to reason, but often hears wrong, like a hasty servant or a guard dog. He contrasts this with desire, which he says does not obey reason, although it is frequently responsible for the weaving of unjust plots. He also says that a bad temper is more natural and less blamable than desire for excessive unnecessary pleasure. And he claims that acts of hubris never result from anger, but always have a connection to pleasure seeking, whereas angry people act from pain, and often regret it.
So there are two ways in which people lose mastery of their own actions and do not act according to their own deliberations, one is through excitability, where a person does not wait for reason but follows the imagination, often having not been prepared for events. The other worse and less curable case is that of a weak person who has thought things through, but fails to do as deliberated because they are carried in another direction by a passion. Nevertheless it is better to have akrasia than the true vice of akolasia, where intemperate choices are deliberately chosen for their own sake. Such people do not even know they are wrong, and feel no regrets. These are even less curable.

Donald Davidson (1969/1980) attempted to solve the problem by first criticizing earlier thinkers who wanted to limit the scope of akrasia to agents who despite having reached a rational decision were somehow swerved off their “desired” tracks. Indeed, Davidson expands akrasia to include any judgment that is reached but not fulfilled, whether it be as a result of an opinion, a real or imagined good, or a moral belief. “[T]he puzzle I shall discuss depends only on the attitude or belief of the agent…my subject concerns evaluative judgments, whether they are analyzed cognitively, prescriptively, or otherwise.” Thus he expands akrasia to include cases in which the agent seeks to fulfill desires, for example, but ends up denying himself the pleasure he has deemed most choice-worthy.

Davidson sees the problem as one of reconciling the following apparently inconsistent triad:

If an agent believes A to be better than B, then they want to do A more than B.
If an agent wants to do A more than B, then they will do A rather than B if they only do one.
Sometimes an agent acts against their better judgment.

Davidson solves the problem by saying that, when people act in this way, they temporarily believe that the worse course of action is better, because they have not made an all-things-considered judgment, but only a judgment based on a subset of possible considerations.

Another contemporary philosopher, Amélie Rorty (1980) has tackled the problem by distilling out akrasia's many forms. She contends that akrasia is manifested in different stages of the practical reasoning process. She enumerates four types of akrasia: akrasia of direction or aim, of interpretation, of irrationality, and of character. She separates the practical reasoning process into four steps, showing the breakdown that may occur between each step and how each constitutes an akratic state.

Another explanation is that there are different forms of motivation which can conflict with each other. Throughout the ages, many have identified a conflict between reason and emotion, which might make it possible to believe that one should do A rather than B, but still end up wanting to do B more than A.

Psychologist George Ainslie argues that akrasia results from the empirically verified phenomenon of hyperbolic discounting, which causes us to make different judgements close to a reward than we will when further from it.
Much of the philosophical literature takes akrasia to be the same thing as weakness of the will. So, for example, smokers who judge that it is best for them to quit smoking, but don't quit, act against their better judgment and therein display weakness of will. That is, their being weak-willed consists in their failing to do what they think is best.
However, some have challenged the link. Richard Holton (1999), for example, argues that weakness of the will involves revising one's intentions too easily. Under this view, it is possible to act against one's better judgment (that is, be akratic), but without being weak-willed. Suppose, for example, Sarah judges that taking revenge upon a murderer is not the best course of action, but intends to take the revenge anyway and holds to that intention. According to Holton, Sarah behaves akratically but does not show weakness of will.

Another view is that although the person holds certain moral views in high esteem—such as, say, murder is wrong or revenge is wrong—the person holds other beliefs more strongly, such as doling out moral deserts or staying true to one's friends. With this in mind, the moral conceptual framework of the individual must be evaluated to determine the nature of the act. To show strength of will implies a pre-determined decision-making process that may or may not seem to be in conflict with generally accepted moral beliefs.