This has often been claimed, and it is widely accepted that the agency involves the initiation of actions by the agent.  But it was controversial what that was. The standard concept is consistent with the claim that intentional actions are initiated by the agent, and proponents of the standard theory have argued that initiation can be explained by the agent`s mental states and events. According to the pious wish versions of the point of view, initiation by the agent consists of causality by the relevant wish-faith pairs (Goldman 1970; Davidson, 1971; Dretske, 1988). According to recent versions, initiation consists of causality by the corresponding intentions (Brand 1984; Bratman, 1987; Bishop, 1989; Mele 1992, 2003; Enç, 2003). However, opponents of the standard design argue that an agent`s ability to act cannot be reduced to the ability to act intentionally and for reasons. They argue that the exercise of free will can be completely spontaneous, in the sense that an agent can initiate an action without reason and without prior intent. From this point of view, reasons and intentions can have a strong and even decisive influence on how an agent acts. But the ability to act has its source in the power to initiate, and the exercise of this power cannot be reduced to the fact that the agent is displaced for reasons or intentions.
This is an alternative conception of the Agency (Ginet 1990; O`Connor, 2000; Lowe, 2008; see also McCann, 1998; for a critical discussion, see Mele 2003: 38-51, 71-76; Clarke, 2003: 17-24). Proponents of this alternative conception reject standard theory and generally reject any representation of the ability to act in relation to causal relationships between states and events implied by agents. According to some, the initiation of actions consists of an irreducible causality of pathogens, others appeal to mental acts not caused by the will. The main positions on this subject correspond to the main positions of the Agency`s metaphysics, to which we refer in section 3.1. However, if we look sociologically at the daily lives of people among disenfranchised and oppressed populations, we see that freedom of choice is alive and well and takes many forms. For example, many take the life course of black and Latino boys, especially those born into lower socio-economic classes, as largely predetermined by a racy and classified social structure that imprisons the poor in neighborhoods without jobs or resources, places them in underfunded and understaffed schools, persecutes them in remedial classes, and disproportionately monitors and punishes them. Yet, despite a social structure that produces such troubling phenomena, sociologists have found that black and Latino boys and other disenfranchised and oppressed groups exercise their free will in various ways in this social context. It may seem obvious that our mental life is filled with mental actions. We participate, consider, judge, argue, advise, accept, support, decide, try and so on. It may seem like these are all things we do. If we look at such cases through the standard theory of the ability to act, we immediately encounter two difficulties.
First, it seems that such mental events are almost never, if ever, intentional actions. According to the standard theory, an event is an intentional action of type A only if the agent has an intention that includes A in its content. In the basic case, this would be an intention on A. In an instrumental case, it would be an intention to perform another action B to reach A. Now, thoughts are partly individualized by their content. Take the thought that p. According to standard theory, thinking that p is an intentional action only if the agent has an intention that includes „thinking that p“ in its content. This is quite strange and problematic because one would have to intend to think a certain thought before thinking about it. Secondly, there are problems with the central case of decision-making. According to the standard theory, the decision for A would only be a deliberate action if one already intended to make a decision that includes the „decision for A“ in its content.
This, in turn, seems quite strange and problematic. Also, our reasons for making a decision for A are usually our reasons for A – these are reasons for carrying out the action. According to standard theory, something is an action only if it has a basic explanation (in relation to the agent`s desires, beliefs, and intentions). Since reasons are usually reasons for action, it is still difficult to see how a decision can be an action. Considerations of this kind may lead to the conclusion that thoughts are almost never, if ever, mental actions (see Strawson 2003). So, first, the agency as free will. This one is pretty simple, I think, but more common than you think. I think it all goes back to Durkheim and his presentation of the sociological agenda in Rules. I have a page number somewhere where he defines his project in such a way that he examines the limits of social facts about people`s free will. If I remember correctly, Durkheim only uses the term free will. Later authors invoke free will, but they often mean something more philosophical: the sacred power of an individual to be an individual, to have goals that are somehow not caused by social forces, and to reshape the world to achieve those goals.
I think this particular definition of agency is particularly useless for sociology, even for sociology, which is not centered on scientism. In all our struggle to show how different aspects of social life are constructed, down to the individual and individualism (cf. Marx, Foucault), we must not allow an essentialized idea of individuals to take a back seat so easily. In the broadest sense, the agency is practically everywhere. Whenever entities enter causal relationships, it can be said that they interact and interact with each other, leading to changes together. In this very broad sense, it is possible to identify agents and agency, as well as patients and compatibility, virtually anywhere.  However, as a general rule, the term „agency“ is used in a much narrower sense to refer to the performance of intentional acts. This way of thinking the agency has a long history in philosophy and goes back to Hume and Aristotle, among others. In contemporary analytic philosophy, he is most often associated with the influential works of Anscombe (1957) and Davidson (1963).
Anscombe and Davidson`s views differ considerably in many ways, but they share the central lesson that action must be explained in terms of the intentionality of intentional action. In the ensuing debates, the philosophy of action revolved largely around the concept of intentional action. For some time, the term „agency“ was rarely used, and when it was, it was generally understood as referring to the exercise of the ability to perform intentional actions.  This has changed in the more recent debate, in which in many areas of philosophy (and in other areas of research) there is increasing talk of the ability to act.  To some extent, this focus on the agency concept has been fueled by resistance to the agency`s assimilation into deliberate action. . . .