D.P.P. v. Smith--A New Study

I just got back into town and figured I should post something. So, here it is. Recently I ran another preliminary study (part of a much larger study to come!). The cases are based on a real court case--D.P.P. v. Smith (1961). In this landmark case, jurors in England had to determine the guilt of a man named Smith who had driven a car containing stolen goods in a zigzag course in order to shake off a policeman who had been clinging to the side of the car. When the policeman was finally shaken off, he rolled into oncoming traffic and sustained fatal injuries (D.P.P v. Smith [1961] A.C. 290). I wanted to see whether moral considerations affect people's judgments concerning a) whether Smith knowingly brought about the officer's death, and b) whether Smith intentionally brought about the officer's death. So, I developed two cases--one involving a thief and an officer and another involving a driver and a car-jacker.
Subjects were 126 undergraduates--each of whom received one of the following two vignettes:

Case 1:
Imagine that a thief is driving a car full of recently stolen goods. While he is waiting at a red light, a police officer comes up to the window of the car while brandishing a gun. When he sees the officer, the thief speeds off through the intersection. Amazingly, the officer manages to hold on to the side of the car as it speeds off. The thief swerves in a zigzag fashion in the hopes of escaping—knowing full well that doing so places the officer in grave danger. But the thief doesn’t care; he just wants to get away. Unfortunately for the officer, the thief’s attempt to shake him off is successful. As a result, the officer rolls into oncoming traffic and sustains fatal injuries. He dies minutes later.

Questions:

1) Did the thief knowingly bring about the officer’s death? Yes No

2) Did the thief intentionally bring about the officer’s death? Yes No

3) How much blame does the thief deserve for the death of the officer (On a scale from 0 to 6—0
being no blame and 6 being a lot of blame)?

Case 2:
Imagine that a man is waiting in his car at a red light. Suddenly, a car thief approaches his window while brandishing a gun. When he sees the thief, the driver panics and speeds off through the intersection. Amazingly, the thief manages to hold on to the side of the car as it speeds off. The driver swerves in a zigzag fashion in the hopes of escaping—knowing full well that doing so places the thief in grave danger. But the driver doesn’t care; he just wants to get away. Unfortunately for the thief, the driver’s attempt to shake him off is successful. As a result, the thief rolls into oncoming traffic and sustains fatal injuries. He dies minutes later.

Questions:

1) Did the driver knowingly bring about the thief’s death? Yes No

2) Did the driver intentionally bring about the thief’s death? Yes No

3) How much blame does the driver deserve for the death of the thief (On a scale from 0 to 6—0
being no blame and 6 being a lot of blame)?

The results were as follows:

Case 1:
Q1) Y 46
N 16

Q2) Y 22
N 40

Q3) Y 5.9
N 4.65

Case 2:
Q1) Y 34
N 30

Q2) Y 6
N 58

Q3) Y 4
N 1.89

These results further support the claim that moral considerations--especially negative ones such as moral badness and/or blame--can have a pronounced effect on folk ascriptions of intentional action. More interestingly, this data suggest that moral considerations may act expansively on ascriptions of knowledge or forsight as well. Indeed, it is this latter possibility that I will explore in greater depth in the study I will be running in a few weeks. For now, I want to pose the following question: If moral considerations have a predictable biasing effect on folk ascriptions of intentional action (and the data suggest that they do), does this create a problem for jury biasing in cases where jurors are asked to rely on their ascriptions of intentional action in determining a defendant's guilt (e.g. in the Smith case)? If so, what is the proper remedy? If not, why not?

Comments

[Neil]

The cases seem unnecessarily confusing to me. Clearly in both cases the intention of the driver is to escape. He does not intend to kill, nor does he believe that killing the cop/thief is a necessary condition of escaping. He may not even foresee that there is a real risk of death (your "placing in grave danger" is ambiguous here). Why not make the focus harm, rather than death, as in earlier cases involving harm to the environment? The results would be no less significant, it seems, and these complications would be avoided.

[john doris]

I wonder if the effect of moral considerations need in every case, or even very often, be thought of 'biasing.' An alternate reading is that ascriptions of intentionality quite properly have a normative element.

--jmd

[Eddy Nahmias]

John's remark really gets at the central issue of these studies on folk intuitions about intentional action (and perhaps other such studies): Supposing that they do uncover certain elements of our ordinary beliefs and concepts, and supposing that these elements conflict with a philosophical theory or analysis of that concept, should the philosophical theory bend to account for the folk intuitions or should philosophers come up with an error theory to explain the intuitions as a mistake?

If you imagine that the Haidt etc. intuitionist view is roughly right across the board, and people's moral judgments are largely the result of emotional (and intuitive) reactions to cases, with reasoned judgments then following these initial reactions in order to justify or explain them, then we need to decide whether these emotional reactions are a better guide to the 'truth' about moral issues or whether they offer an error theory as to why such reactions are mistaken. If we take a pessimistic(?) view and think philosopher's theories are attempts to justify the philosopher's intuitive reactions, then we have an interesting error theory for a lot of philosophical work and philosophical stalemates.

[Neil]

I think we need to be careful in circumscribing the skeptical thesis to which these cases might be seen to lend support. At very least, we seem to need to distinguish between ascriptions of intentionality and intentionality. Suppose intentions are mental entities with neural correlates; then these experiments can't show that intentions have a normative element (of the kind suggested). If we could define and detect intentions independently (eg, by the use of fMRI machines, or Libet-style readiness potential correlations) we could begin to discover whether there are better and worse methods of intention ascription. We might be led to skepticism about, say, a jury's ability to ascribe intentions accurately, but perhaps not about all methods. On the other hand, perhaps intentions are not mental entities at all, but merely instrumental; in that case our scepticism might reach as far as the existence of intentions.

[tnadelhoffer]

Neil--Thanks for the helpful comments. The reason I used "intentionally bring about the death of" rather than "intentionally harmed" is that I was trying to get some insight into whether moral considerations affect juror's judgments about intentional action. And since in the original Smith case (which was quite interesting given that the jury instructions relied on an objective--rather than the subsequently used subjective--standard for ascriptions of intentional action) Smith was charged with the American equivalent of first degree murder--which requires that the agent either intended to kill the officer or intentionally did so. Unfortunately, the law is not always very clear on the distinction between intending to x and intentionally x-ing--a problem that I will address in some detail in my dissertation.

John--The reason I have called the effect that moral considerations have on folk ascriptions of intentional action a "biasing effect" is that given two very similar cases--with moral considerations (or lack thereof) being the sole difference--subjects will make very different judgments. So, for instance, if one group of subjects is asked whether an agent intentionally roled a six in an effort to win a game and another group is asked whether an agent intentionally rolled a six in an effort to set off some kind of James Bond inspired murder machine, significantly more of the subjects in the second group will judge that the agent x-ed intentionally even though the cases are identical in terms of control, skill, foresight, etc. So, I am unsure what else to call this effect--even if I agree that "biasing" may be a bit more pejorative than I intended. These considerations notwithstanding, I am still unsure what to make of your suggestion that perhaps "ascriptions of intentionality quite properly have a normative element." Indeed, you are not alone in making the suggestion. R.A. Duff, Bratman, Harman, and Knobe seem to argue along similar lines. Nevertheless, there are seemingly innumerable mundane instances of intentional action that on the surface do not appear to involve any normative element at all. Take, for instance, the difference between an agent's accidentally tripping on the one hand and her intentionally tripping in an effort to make a young child laugh on the other hand. Of course, you could suggest (and I would be inclined to agree) that part of what it means to say that something was done intentionally is to say that the agent is (or could be) responsible for it. On this view, we would say that in the case of the accidental tripping, if the agent happened to make a child laugh, she would not deserve any credit for having done so--whereas in the case of the intentional tripping, the agent would deserve some credit/praise for having made the child laugh. Is this the sort of thing you have in mind when you say the ascriptions of intentional action may have a normative element built in? If not, I would be interested in hearing your take on it--especially given that it is directly relevant to my dissertation!

[Paulo Sousa]

Hi Thomas,

Here is a comment about the implications of your results to jury biasing. The use of the adverbs 'knowingly' and ‘intentionally’ may not correspond to the technical definitions of ‘knowingly’ and ‘intentionally’ in the context of the definition of different levels of mens rea (When the legal system clearly make a distinction between these two levels – for example, in American jurisdictions that follow the Model Penal Code –, this means simply that “someone is practically certain that his conduct will cause a result” and that “it is someone’s conscious objective to cause a result”, respectively.) I don’t think that juries receive instructions framed in the way you phrase your questions - they are also given some more precise definition of what "knowingly" or “intentionally” means in the context of the mens rea of a particular offense. To show implications for jury decisions, you have to phrase your questions accordingly – e.g. "Was X practically certain that by shaking Y off he would cause Y's death?” Also, in the context of American Culture, the discussion prompted by your first scenario seems to be completely irrelevant to jury biasing, since the person can be accused of first degree murder by a charge of felony murder, the definition of which is independent of any mental state related to the death result.

[Eric Rovie]

Hey Tom,

Just a thought...is there already some biasing going on in the description, in case 2, of a "car thief"? Maybe that's the point of the description. But it might be less obvious if the description was of a person in a police uniform (in case 1) and of a stranger with a gun (in case 2). Of course, the former could be a car thief in a uniform and the latter an undercover cop, but the point seems to be that the MORAL assessments might be riding hard on the descriptions of "police officer" and "car thief" and not on the other details. Again, that might be your point, I'm not sure. Regardless, it's interesting stuff.

[Tom Morris]

It's worth reading the original case. Here's an excerpt: http://www.lawrence.edu/fac/boardmaw/ddp_v_smith_Br_lords.html

[Dave Gudeman]

I suspect that what you are really measuring here is the mutability of words in context. The words "knowingly", and "intentionally" will tend to take on the meaning "blamefully" in a context like your story. I would expect the answers to those questions to reflect the answer to the blame scale for some unknown percentage of respondents.

You could try to correct for this by carefully defining what you mean by the words, but you would have to do it without examples that might poison the results.

(I didn't understand your summary of the blame scale. It's a 0-6 question but you give Y/N summaries).

[Dave Gudeman]

Sorry for the double comment but I just realized. Another way to avoid the problem of mutable meaning is to phrase the questions in a more common way:

1. Did the X know that what he was doing would kill the Y?

2. Did the X want to kill the Y?

[tnadelhoffer]

Dave,
Thanks for the comments. I can see why the way I reported the blame ratings looks funny--the number after the Y is the average blame rating of participants who said that Smith intentionally brough about officer's death and the N is the average blame rating of those who said he did not. Of course, I should have included the overall blame rating, but that would have obscured the fact that those who did not think he did it intentionally found Smith less blameworthy--which is worth emphasizing even if it is unsurprising. As for your comments about the "mutability of words in context": finding a way of that does not either "poison the results" or beg other interesting questions (e.g. the view that if you wanted to do x and you did x, you intended to do x--or the view that if you knew that you would bring about y by doing x, you both intended to y and you intentionally brought y about) is quite difficult. So, I appreciate your suggestions. Your point is well taken--indeed, Fred Adams has taken this line of reasoning in response to some of Knobe's studies (although he talks in terms of pragmatic implicature). Let me know if you find the recent studies lI have posted to be less problematic in this regard (i.e. the water pumper and lawn mower vignettes). I will also be posting some more Smith studies, as well as, some studies based on Bentham's stag-hunter example in a few days. I welcome your comments.

[larry nahmias]

The first scenario represents the reaction of the thief as an attempt to avoid prosecution and imprisonment while the second represents the reaction of a man to an unknown person approaching the car with a gun, and continuing to hold on (presumably with the gun) as the car speeds away. The man does not know the person with the gun is a car thief. One scenario is an escape from responsibility, the other an escape, presumably, for one's life. Because of this, I don't agree that the scenarios equally measure moral considerations.

[shadia]

Hi,
I don't understand the way you collected and gathered the data u found for the third question using the 0-6 rating scale. Did u use some kind of criteria that grouped responses to that question together? Like answers between 0-2 means so and so? and furthermore was it the same method employed by joshua Knobe in his work on intentional action and side effects in ordinary language. Please let me know asap. Thanks.