III

As intentions are so bound up with agent putting forth of efforts, especially physical doing, Anscombe directs her discussion to how intentions related to actions. There are two kinds of cases for intentions to relate to action, intentional action and intention with which an action is done. In discussing intentional action, Anscombe distinguishes events that are intentional from events that are not. From the agent’s perspective, events that are intentional renders a special sense of the question “why” applicable and, in turn, the answer to “why” provides “reason for acting.” On the one hand, the question “why” addresses the fact that an intentional action usually has a reason when the action is examined retrospectively. “Why did you knock the cup off the table?” – “I was startled by the look on his face.” On the other hand, the question “why” addresses the fact that intentional action usually has a purpose prospectively. “Why are you taking a taxi?” - “Because I want to go visit my grandmother.” When an intention and an action are looked at concurrently, the “why” question provides a reason for acting. “Why are you staring into the sky for no reason?” - “Because three other people are doing it.” The special sense of “why” question all find application in these intentional actions.

For actions not are not intentional, the special sense of “why” question is rendered inapplicable. “Why did you do that?” – “I did not know that I was doing that.” Or “Why are you doing this?” – “I am not aware that I am doing this.” The inapplicability of the “why” question here, according to Anscombe, is similar to the question “How much money do you have in your pocket?” – “I have none.” Since an action can have multiple descriptions, one may know that he is intentionally performing an action under one description but not under another. “Why are you making so much noise?” – “I am not aware that I am making noise. I am practicing my new piano piece.” Or “Why did you shut me outside of the house?” – “I did not mean to shut you out of the house. I wanted to shut the door for the night.” Or “Why did you break my Lego construction?” – “I only tried to break the Lego construction in front of me. I did not know I broke yours.” Therefore, for an action that has multiple descriptions, only the ones that the agent is aware of are the descriptions for the intentional action. The action under the descriptions that the agent is not aware of has no application for the special sense of the “Why” question.

Using the special sense of the “Why” question as the criteria for determining whether an action is intentional may run into difficulties. For example, “Why are you lying down?” – “I am just lying down for no reason.” In this case, the agent is aware of the action, but has neither a reason for it nor a purpose for it. Certain actions are involuntary – yawning, sneezing, muscle spasm, etc. and these are clearly not intentional actions. However, one could fall victim to moments of irrationality. When someone becomes extremely angry with another, even though he knows not to shout or scream, even though he takes time beforehand to think about what his reactions ought to be and concludes with certainty that he ought not to shout or scream, he nevertheless starts to shout as soon as he see the person he is angry with. Although he is not insane at that moment of shouting and is aware of what he is doing, yet he continues to shout. “Why am I shouting?” “I can’t help it.” Is his shouting not an intentional action? It would be counterintuitive to decide that such cases of performing irrational actions against one’s better judgement despite his awareness are not intentional actions.

Anscombe includes such actions as intentional actions through discussion of mental causes for intentional action. For example, “why did you shut the door in such a panic? – “I saw a bear coming this way and was startled.” Such a cause, from internal and external stimulation, and not arising from a desire, is a mental cause. The question earlier “Why am I shouting?” should be more accurately answered “Because I am angry.” Anger is the mental cause for shouting. A mental cause for an intentional action is not a mental event. The smile of the girl on the street reminds me of my daughter and I can’t help but taking a step toward her. It is not a mental event of desire or belief but a mental cause due to sensory input and stimulation. Anscombe says: “a mental cause is what produced this action or thought or feeling on your part: what did you see or hear or feel, or what ideas or images cropped up in your mind, and led up to it?” ⁴

Anscombe then distinguishes between motives and intentions. For example, if a man intends to kill someone, he could do out of a good motive or a bad motive. As Anscombe describes, he could do the killing in the forms of “to release the victim from the awful suffering” or in the forms of “to get rid of the swine.” As Anscombe describes that there is such phenomenon as doing A in order to B, doing B in order to C, doing C in order to D, and so forth. Here, the action of killing is A, and B stands in for either releasing the victim from the awful suffering or getting rid of the swine. A, B, C, D, are in some ways either chain reactions or ripple effects. Anscombe may call killing an intention and releasing the victim from the awful suffering or getting rid of the swine motives. I do not wish to make such distinctions. A, B, C, D, if conceived in the agent’s mind, if foreseeable, they are all intentions. Perhaps one is in closer proximity to the action, and the other requires additional conditions to come to pass. But if foreseeable, all of A, B, C, D are in the agent’s mind, they are all intentions with which the agent act. When Harry Truman signed on a piece of paper to authorize the bombing of Hiroshima and Nagasaki, his intention was to end the war. His action also cost the lives of thousands of innocent civilians. For that, Anscombe put Truman in the league of Genghis Khan, Nero, or Hitler and opposed Oxford University’s proposal to give Truman an honorary degree. ⁵ Perhaps Truman had no intention to hurt innocent civilians on its own. But the action of signing on the piece of paper carried the intention of ending the war and the intention of killing innocent civilians because while the intention of ending the war was what was sought after, the killing of innocent civilians is so obvious an effect that an intention must be constructed. If one insists that some certain and obvious result of an action does not occur in one’s mind prior to the action, he will be charged with willful blindness. Willful blindness is a way of objectively constructing an intention when the agent denies having the subjective intention. The construction of intention says that the result of the action is so obvious and natural to someone of your capacity that for you not to see the result is inconceivable. There is also the phenomenon of mixed intentions. Perhaps two people are in a good relationship and they love each other. But one of them needs a green card in order to stay in the United States. So, when they get married, they have the dual intention of being committed to sharing their lives together and helping one of them to obtain a green card so that they can remain together. Such issue of dual intention posts no conflict in the framework we discuss thus far.

Perhaps there is a different way to identify an intentional action. Perhaps we check whether there is a counterpart in the agent’s mind that corresponds with the action being performed. Let’s say an intentional action ought to have a counterpart in the mind. For example, if I have an intention to move the table and I am moving the table, then the action of moving the table is intentional because the action has a counterpart in my mind. In contrast, assume I knock over the chair while moving the table, knocking over the chair is not an intentional act because the action of knocking over the chair does not have a counterpart in my mind while the knocking over is happening. It seems that, in order for an action to be intentional, the counterpart in mind has to be time bound. An intention occurs after the action is not a counterpart. There must exist a kind of mapping between the counterpart in mind and the action prior or at the time of the action. What does a mental counterpart look like? Certainly, the counterpart and the action are not identical. For example, if someone intends to cross a river on foot but doesn’t know where his steps should land, every step is a trial and error and it is impossible for him to know how to take each step beforehand. It may then be impossible to have the kind of accurate counterpart in mind to map to the action. However, as long as mapping between intention, i.e., the counterpart in the mind, and the action is possible, symmetry between the two is not necessary.

Anscombe also characterizes an intentional action as something that the agent knows through “knowledge without observation.” This makes sense because an intentional action is one where there is a counterpart in the mind. Knowing this counterpart in mind, the intention, is part of self-knowledge. As Akeel Bilgrami convincingly argues in Self-Knowledge and Resentment (2006), an agent has direct access to self-knowledge due to the self-knowledge’s special property of transparency. Also, as Bilgrami argues, this self-knowledge of intentional state also has the property of authority – as discussed earlier, the person with the intention has the ultimate authority to settle what the intention is. It occurs to me that, with an intentional action, self-knowledge of one’s intention and performing the action according to intention are insufficient to make the whole process complete. The agent must also know and confirm that the action being performed is indeed the action intended. In other words, to have an intentional action, the agent must also have knowledge of the external world and ensure that there is correct mapping of what is in the external to what is in the internal. For example, if someone intends to cut a tree, but he instead cuts the family dining table, believing that he is cutting the tree. This is not an intentional action based on our earlier discussion requiring that there exist a kind of mapping between the counterpart in mind and the action at the time of the action. There is clearly no mapping between the action of cutting the family dining table and what’s in the person’s mind. Looking at it from the angle of knowledge, although the agent has self-knowledge of his intention, he nevertheless lacks knowledge that his action is not in accordance with this intention. What is internal to the agent is the agent’s first-person perspective. What is external is an objective third-person perspective. Therefore, intentional action requires that the agent performs a matching of a first-person perspective and a third-person perceptive. ⁶

Perhaps not all intentional actions form a sub-class of the things a person knows without observation. Although the intention is knowable to the agent due to the transparency property of self-knowledge, an intentional action, if it interacts with any external object, requires the agent to make an observation to ensure that the action is indeed as intended. It is true that we intuitively understand that certain intentional actions do not require observation, for example, if I intend to lift my index finger up from the computer keyboard, I could do so without observation. I could move my index finger to touch my tip of the nose with my eyes closed, like the sobriety test performed by a police officer on potential drunk driver. These are actions that we intuitively come to understand that a well-functioning human being is capable of performing. However, actions that involve an external object must require the supplemental knowledge of the external object through observation. If I intend to sit down on a chair, I must make observations to make sure I will indeed sit on the chair. Anscombe raises this problem of intentional actions requiring two objects of knowledge – one’s intention and what is actually taking place.7 Because Anscombe has argued that one’s knowledge of what one does is not by observation, her solution for this problem includes the same argument. Her formula for addressing the problem is what she refers to as “I do what happens.” When the description of what happens is the very thing which I should say I was doing, then there is no distinction between my doing and the thing’s happening. 8 This explanation is not entirely satisfying. Not all intentional actions are like throwing rocks out of the window – you intend to do something, take your action and accept what happens. Some intentional actions require constant negotiation with the environment, e.g., crossing the river one step at a time on foot, and therefore requires external observation to continue to carry out the intention.

I would like to suggest that knowledge of one’s intentions, or self-knowledge, is most meaningful and useful when the self-knowledge also contains knowledge outside of one’s own mind, i.e. knowledge by observation. When one suddenly wakes up in the middle of the night, she knows it is night time. She believes that it is night time – self-knowledge. However, her self-knowledge would be more meaningful and useful if she checks out the clock on the night stand to find out that it is 2:30 AM. Similarly, if one is dropped off by the road side on the way from New York to Washington, DC in the dark of the night, self-knowledge may give him the idea that he is on the way from New York to Washington, DC. But the self-knowledge will be more helpful and meaningful if he could rely on his GPS to find out that he is only five-minute walk away from a train station. Does needing external observation cause self-knowledge of the intention to lose the property of authority – the agent has the final authority on the content of the intention? I do not think so. If something has both a private element and a public element, the agent as the only one who has access to the private element will remain the final authority on what the something is.