The anatomy of intentional action

In recent years, cognitive scientists have been trying to explain a curious psychological quirk called the ‘Knobe effect’. A new paper by Chandra Sekhar Sripada and Sara Konrath in Mind & Language offers an answer.

(As this is a long post, I’ve formatted it as a PDF so you can print it out and read it over a coffee if you prefer)

A little under a decade ago, a young cognitive scientist called Joshua Knobe, now at Yale University, ran a series of simple experiments with people in a Manhattan Park. He asked half of them to read the following story:

The vice-president of a company went to the chairman of the board and said, ‘We are thinking of starting a new program. It will help us increase profits, but it will also harm the environment.’ The chairman of the board answered, ‘I don’t care at all about harming the environment. I just want to make as much profit as I can. Let’s start the new program.’ They started the new program. Sure enough, the environment was harmed.

He then asked his participants, “Did the Chairman intentionally harm the environment?”. The vast majority — around 82% — said yes, he did.

Meanwhile, Knobe asked the other half of his interviewees to read exactly the same passage, but with the word ‘harm’ replaced with ‘help’, and then had them answer the question, “Did the Chairman intentionally help the environment?” Now most people (77%) said no, he did not1,2.

These results — and they’ve been replicated many times since — are a bit surprising. The mechanics of both situations are identical: the chairman signs off a project irrespective of its environmental impact, although in one scenario the outcome is bad (the environment gets harmed), and in the other, it’s good (the environment is helped). So shouldn’t the outcomes, whether good or bad, be seen as equally intentional or unintentional? Well, these findings suggest they’re not: it looks like the ‘moral flavour’ of the outcomes of actions — that is, whether they’re good or bad — shapes whether or not we construe those actions as having been performed intentionally.

Yet it’s natural to think that our moral judgments about the outcomes of people’s actions are themselves sensitive to whether or not those outcomes were intended or mere side effects. There’s a world of difference between me walking up to someone in the street and punching them square in the face, and accidentally hitting someone in the face as I’m trying to provide directions to another passerby (a feat I once achieved on London’s Tower Bridge). In the first case, my intentions would be bad, and so the outcome is judged as a moral offence; the latter was an accident, a side-effect of trying to help someone else, and therefore not morally blameworthy. It seems that judgments about intentionality should come first, with the moral judgment deriving from that.

In the scenarios Knobe explored, however, the influence sometimes goes the other way: when outcomes are bad, the agent whose actions lead to those outcomes is judged to have acted intentionally, even if the outcomes are technically side effects. But this isn’t true when positive outcomes, such as helping the environment, emerge as side effects of actions taken for other reasons, like pursuing profit. This asymmetry in judgments about intentionality is now well-known as the Knobe effect.

How to explain it? Perhaps the most obvious solution turns on the moral quality of the outcomes in question. When they’re good, and arise as side effects in the pursuit of other ends, they’re seen as just that: unintentional side effects, not the sort of thing to praise. But when the same actions lead to bad outcomes we apply the concept of intentionality differently, and treat analogous side effects as being intentionally produced. This helps us make sense of blaming people for the bad outcomes they create, whether on purpose or not.

Alternatively, ascriptions of intentionality may turn not so much on outcomes per se, but on judgments about the moral status of the person behind the relevant actions. If they violate a moral norm, by harming the environment, then we see them as immoral. This causes us to switch to a ‘moral blame’ mode, which makes us more likely to say that they acted intentionally — just as most most people do in the Chairman’s harm scenario.

These slightly different explanations are both, however, rooted in judgments about good and bad, right and wrong (whether they be directed at outcomes or agents). In the philosophical lexicon, they’re based on ‘normative factors’. And normative explanations of the Knobe effect are currently the most popular.

What might link normative judgments with intentionality judgments? One possibility is the negative emotions that are typically aroused when we hear about people doing things that lead to bad or immoral outcomes. Perhaps these feelings feed into how we deploy the concept of intentional action, so that perceived violations of moral norms — like harming the environment — are treated as intentional outcomes, even though they’d be seen as unintentional side effects if the outcomes were positive.

Unfortunately, there are reasons to believe this isn’t the case. The evidence comes from patients with damage to brain regions associated with processing negative emotions. These patients, who have sustained damage to the ventromedial prefrontal cortex (VMPFC), are not impaired intellectually, and have participated in a number of studies on moral cognition, in which they show a markedly different pattern of moral judgments compared with the general population.

Take the well-known Footbridge dilemma: an out of control train is on course to hit and kill a group of five walkers stuck on the tracks ahead, and the only way to stop the train is to push a very large and heavy man off a footbridge over the tracks into the train’s pain, stopping the train but killing the man. Is it morally permissible to push the man to his death to save five others? Most people say no, and their response in this and similar cases is associated with greater activity in the VMPFC, suggesting that the emotional aversion to harming another person inhibits this choice. The brain-damaged patients, by contrast, typically favour more a more utilitarian — some might say colder — logic, and the majority say it is OK to push the man. Yet despite this deficit in processing negative emotions, they still show the Knobe effect, indicating that it does not depend on these emotions3.

These findings do not mean that the normative theories of the Knobe effect are wrong; it just means they have to come up with another mechanism apart from the negative emotions elicited by violations of normative standards to explain the Knobe effect. In any case, normative theories, while very popular, are not the only game in town. Another contender has been developed by Chandra Sekhar Sripada4 (co-author with Sara Konrath of the new paper5 to be discussed shortly). Sripada’s account has the imposing title of the deep-self concordance model (DSCM), and it takes a second to get your head around. But it’s a neat model, so it’s worth it.

The basic idea is that when we encounter cases like Knobe’s Chairman scenario, we make two different judgments: one about his attitudes and values — is he anti- or pro-environmental? — and another about whether the outcomes of his actions are consistent with those attitudes. Neither of these judgments is normative; they are descriptive judgments about the facts of the case. In both scenarios we’ve been looking at, the Chairman reveals an anti-environmental attitude: he says he “doesn’t care about the environment”. (Though he doesn’t explicitly say that he’s against the environment and wants to cause environmental damage, his indifferent attitude is the sort leads to anti-environmental activities.) In the case where the project he approves harms the environment, the outcome is concordant with his anti-environmental stance, and so the outcome is seen as intentional. However, when the project has pro-environmental consequences, these are seen as discordant with his attitude, and are therefore not considered to be intentional outcomes.

In short, the normative explanations claim that ascriptions of intentionality are driven by judgments about the rightness and wrongness of actions and outcomes – that is, on evaluative judgments about the moral status of agents and their deeds. The DSCM, by contrast, focuses on descriptive judgments about the evaluative attitudes that agents hold.

A real-world example of the difference between forming evaluative judgments of an agent, and attributing evaluative attitudes to an agent, will bring out these distinctions a little more clearly. Let’s say George W. Bush gives a speech about abortion. Listening to this, we’re likely to make two distinct kinds of judgment: what we think the content of his speech is (the descriptive aspect), and how we feel about that content (the evaluative, normative aspect). Specifically, we’ll come to a view as to whether Bush advocates pro- or anti-abortion sentiments: we will ascribe certain evaluative attitudes (pro- or anti-abortion) to Bush based on descriptive facts about what he said in his speech. At the same time, we’ll decide whether his moral views on abortion are themselves moral or immoral — that is, we’ll make a normative judgment about the evaluative attitudes that we ascribe to Bush.

Notice that there’s likely to be much greater agreement about the descriptive side of this equation. Most people who hear Bush talk about abortion will recognise that he’s against it; that’s the evaluative (normative) attitude most people will ascribe to Bush. Yet there’s much greater disagreement about whether this makes Bush moral or not, as this judgment depends on the separate normative evaluation of Bush (based on the normative attitudes we ascribe to him). As descriptive judgments about the content of other people’s evaluative attitudes often go hand-in-hand with evaluative judgments of those very same people, they can seem like the same kind of judgment, but it’s clear that they are not.

To reiterate: normative theories suggest that whether we say someone did something intentionally or not depends on evaluative (normative) moral judgments about the agent, his behaviour or the outcomes it produces. The DSCM, on the other hand, proposes that judgments of intentionality depend of descriptive judgments about the evaluative (moral) attitudes of other people, and — again descriptively — whether their behaviour and the outcomes produced are concordant with those attitudes: if they are, the action is judged to be intentional; if not, then unintentional. And so this is what Sripada and Konrath set out to test.

To do so, they ran the original Chairman case by 240 study participants: half read the ‘harm’ version, and half the ‘help’ version. Participants then answered a series of question, each designed to tap into different normative and descriptive aspects of the Chairman’s behaviour. The questions were as follows, and each was answered using a seven-point sliding scale:

1. How much do you agree with the statement “The Chairman intentionally harmed [helped] the environment”? (Scale ranged from ‘strongly agree’ to ‘strongly disagree’.)

2. In your view, how good or bad is the outcome that the environment is harmed [helped]? (‘Very bad’ to ‘very good’.)

3. In your view, what is the Chairman’s moral status? (‘Very moral’ to ‘very immoral’.)

4. What are the Chairman’s values and attitudes towards the environment? (‘Very pro-environment’ to ‘very anti-environment’.)

5. In the vignette above, the Chairman’s action brings about an outcome in which the environment is harmed [helped]. In your view, to what extent is the Chairman the kind of person who will, in other contexts and situations, bring about outcomes similar to this one? (‘Very likely’ to ‘very unlikely’.)

6. What are your own values and attitudes towards the environment? (‘Very pro-environment’ to ‘very anti-environment’.)

Question 1 of this series simply gauges whether or not people are more likely to ascribe intentionality to the Chairman in the harm condition than the help condition. They did, providing another confirmation of the Knobe effect. The other questions bear on the possible factors that might underlie this effect. Questions 2 and 3 probe normative evaluations of the outcome of the Chairman’s actions, and his moral status, respectively. Question 4 asks a descriptive question about the values and attitudes of the Chairman. Question 5 is a similarly descriptive judgment about the likelihood that the Chairman’s behaviour will generalise to other situations— that is, whether the behaviour in this case is perceived as an aspect of the Chairman’s ‘deep self’, his enduring attitudes and values; or whether it’s seen as a one-off with no relevance to how we view the Chairman’s deep self. (So Questions 4 and 5 constitute the ‘deep self’ variables.) Question 6, which probes study subjects’ own normative attitudes towards the environment, was included to look at the role of these considerations in judging the Chairman. (The order in which these questions were presented was systematically varied between subjects to eliminate potential order effects.)

Sripada and Konrath didn’t simply looking at the subjects’ responses, and try to work out which factors are driving the Knobe effect. Instead, they employed a technical statistical tool called structural path analysis to achieve the same end. I’m not going to try to explain this in detail, but here is the basic idea.

The starting point is that the Chairman case people read about — either the harming or helping scenario — has an effect on intentionality judgments (the Knobe effect). There are a number of mediating factors that could potentially explain this effect: some normative, others descriptive. So the psychological process involved here can be represented as a series of boxes, starting on the left with Case (harm or help), moving right through the various mediating factors, and ending up with Intentionality Judgments on the right. These boxes can then be connected with arrows or ‘paths’ that represent potential causal connections between each box. The crucial point is that structural path analysis enables a pathway coefficient to be calculated for each causal path, which denotes how strong the effect is on the next box (moving rightward). This all makes much more sense when you actually look at a structural path analysis; Figure 1 shows the first example Sripada and Konrath came up with.

Figure 1. First-pass structural path analysis of intentionality judgments

The way that Sripada and Konrath devised Figure 1 is as follows. Firstly, the whole point of this study is to see how reading about the Chairman harming or helping the environment (the two cases that can sit in the ‘case’ box) influences intentionality judgments. And Sripada and Konrath wanted to see which factors mediate this effect. The questions they had subjects answer tapped into four different factors: Goodness/Badness Judgments (of the outcome); Moral Status (of the Chairman); Chairman Values/Attitudes; and Generalizability (whether the Chairman would behave similarly in other contexts). These four potential meditating factors therefore sit between Case on the left, and Intentionality Judgments on the right. And because Case is presumed to affect judgments of these four factors, there are causal arrows leading from Case to the meditating boxes. Likewise, because these four factors are hypothesised to affect intentionality judgments, there are arrows leading from the former to the latter.

What do the numbers on each path tell us? These are the pathway coefficients, which reflect the strength of the effect of each leftward box on the rightward box to which it points. The details of how these are calculated need not delay us here, but in essence the logic is this. Different people read one case or the other, and then they answer a question about how good or bad the outcomes were, say. On the whole, people who read about the Chairman harming the environment will tend to say ‘bad’ or ‘very bad’, and the converse in the other, helping, case. So there will be a correlation between Case (harm or help) and Goodness/Badness Judgments about the outcome. Similarly, we’d expect a correlation for Case and Moral Status, Chairman Attitudes and Generalizability. Each of these four factors may in turn be correlated with Intentionality Judgment. (In addition, there is a path going straight from Case to Intentionality Judgments, which captures the effects of case that are not mediated through one of the other variables.)

Structural path analysis take the correlations in the raw data and, on the basis of the proposed set of interconnections, calculates the pathway coefficients, which for technical reasons are not simply the correlations between each box. Rather, they reflect the causal contribution of one to the next. The triple asterisks on some of the paths denote highly statistically significant effects (p < 0.001); the dotted lines represent insignificant causal connections.

So what does Figure 1 tell us? Let’s move from left to right. The pathway coefficients indicate that Case has a statistically significant effect on all four proposed meditating variables, being particularly high for Goodness/Badness Judgments, Chairman Attitudes, and Generalizability, and lower for Moral Status. At this point, we might well think, “So all factors play an important role in intentionality judgments”.

But let’s move rightward once again, from the four variables to Intentionality Judgment. Now the pathway coefficients tell a different story. The coefficients leading from the normative factors — Goodness/Badness and Moral Status — are in fact very low and statistically insignificant, suggesting that they are not playing a strong causal role in Intentionality Judgments.

However, the pathway  from the Chairman Attitudes and Generalizability boxes to Intentionality Judgment, as well as the direct pathway from Case to Intentionality, make a much more profound contribution. Score 1 for the DSCM.

The model shown in Figure 1 — that is, the specific causal connections leading from Case to Intentionality Judgments — was devised on the basis that all four variables (normative and deep-self) could all play a role. However, structural path analyses make it possible to ask, of the model as a whole, how well it ‘fits’ the data it has to account for. For the model in Figure 1, the fit was ‘good’.

However, good isn’t great, and there might be room for improvement. Thankfully, structural path analysis also suggests which paths may be modified in order to get a better fit with the data. Revising Figure 1 along these lines gives the model in Figure 2 (see above).Figure 2. Revised structural path analysis of intentionality judgments

Notice that in this revised model, the arrows connecting the two normative factors of Moral Status and Goodness/Badness with Intentionality Judgment have been done away with. Instead, a new arrow has been introduced in which Goodness/Badness Judgments have an effect, albeit a small one, on judgments about the Chairman’s Attitudes, which in turn has an effect on Intentionality Judgments.

Sripada and Konrath carried out further analysis of the network of cause and effect shown in Figure 2. Specifically, they calculated mediation effects for the two normative factors in the model, and the two deep-self variables. Case manipulations — whether you read the harm or help scenario — affect Intentionality Judgments, and mediation effects capture how much of this influence is mediated through the various variables. In the model shown in Figure 2, Sripada and Konrath found that the two deep-self variables mediate 55% of the influence of Case manipulations on Intentionality Judgments; just 4% is mediated through Goodness/Badness Judgments, which in this model act indirectly by affecting judgments about Chairman attitudes. (Moral Status, as it was not connected to any downstream boxes, does not mediate any effect on Intentionality Judgments in this model.)

Statistically, this model has an almost perfect fit with the data, further bolstering the deep-self concordance model, and relegating normative factors to at best a minor causal role in generating the Knobe effect. If the DSCM model is correct, or at least on the right track, then it undermines claims for the pervasive effect of moral considerations on what is construed as intentional action. When it comes to judging whether someone produced an outcome intentionally, it isn’t morality that matters, but consistency: specifically, the concordance between a person’s values and attitudes, and the outcomes of their actions.

Sripada and Konrath didn’t stop here. They also looked at how people make sense of the Knobe effect, by getting a further batch of participants to read both the harm and help scenarios, followed by a note explaining that researchers have demonstrated an asymmetry in intentionality judgments in these cases (in slightly more user-friendly language!). They were then asked what might explain this effect. The vast majority cited normative factors — precisely the explanatory factors that the structural path analyses undermine.

Why are people so introspectively inaccurate in this case? The fact is that our introspective powers frequently prove to be limited, and we often have little access to the unconscious and automatic processes that shape our decisions and judgments. Instead, we typically latch on to the most salient and accessible features of a problem, and cite those as playing the key role in our thought processes – whether or not they really are playing the causal role we assign to them. Sripada and Konrath propose that in the case of the Chairman scenarios, the normative factors — the fact that he did something that led to bad outcomes, and could be an immoral person — are highly salient, and easy to bring to mind. The ideas underlying the deep-self concordance model, by contrast, are much less salient and introspectively accessible. If this is right, it would not only explain why regular folk can be misled as to what’s important in the Chairman case, but also why so philosophers have focused on normative factors in explaining the Knobe effect – they are human, after all!

1. Knobe, J. Intentional action and side effects in ordinary language. Analysis 63, 190–94 (July 2003).

2. Jones, D. The good, the bad and the intentional. The Psychologist 22, 666–669 (2009).

3. Young, L., Cushman, F., Adolphs, R. et al. Does emotion mediate the relationship between an action’s moral status and its intentional status? Journal of Cognition and Culture 6, 265–278 (2006).

4. Sripada, C. S. The Deep Self Model and asymmetries in folk judgments about intentional action. Philosophical Studies 151, 19–76 (2010).

5. Sripada, C. S. & Konrath, S. Telling more than we can know about intentional action. Mind & Language 26, 353–80 (June 2011).


About Dan Jones

Dan Jones is a freelance science writer
This entry was posted in Deep-Self Concordance Model, Knobe effect, moral judgment, Uncategorized. Bookmark the permalink.

34 Responses to The anatomy of intentional action

  1. Tomkow says:

    Notice that the author of the post feels it necessary to explain at some length the relevant sense of “intentional”. And rightly so since, of course, even a sophisticated audience might need clarification on this score. And yet we are supposed to take subject’s untutored responses to the blank question “Was it intentional?” as reflecting the naïve subject’s “intuitions” about ‘intentional’.

    • Dan Jones says:

      Hey Tomkow, thanks for the being the first commenter of this fledgling site! I took a look at your own, and dig the post on trolley problems – people, check it out for the cool illustrations of each moral scenario (and, of course, the accompanying commentary).

      I’m not entirely sure what the problem is with probing people’s naïve, untutored responses to questions about whether people acted intentionally or not, as the point is to find out what conception of intentionality people seem to naturally deploy. I guess to see why this might be something anyone might want to do requires backing up a bit and asking “What’s the purpose of experimental philosophy of this kind?”.

      Here’s my take. Philosophers can sit around and debate what counts as intentional action – just as they might about what consciousness is, or what makes some actions morally permissible and others not. Many times, arguments about what constitutes, say, intentional action are supported with examples which are supposed to be clear instances of intentional action versus those that are clearly unintentional. But clear to whom, exactly? Well, principally people who will read the articles on this topic, which is mainly other philosophers. That is, arguments about intentional action (and other topcis) often pump the intuitions of the philosophers to which these arguments are directed. They may or may not elicit the intuition the argument’s framer expects. But regardless of whether they do or don’t, we can still ask, “Has this got anything to do with what regular people think about intentional action – or are we just talking about the views of a frankly atypical bunch of people?” (Atypical in that they spend a lot of time deep in rigorous conceptual analysis.)

      And that’s why you might want to see how regular people talk about intentional action – and how it relates to moral condemnation, praise etc. This, then, is essentially social psychology, though of a philosophically flavoured variety. Maybe regular folk have a straightforward or neat notion of what counts as intentional action. Perhaps people will only say an outcome is brought about intentionally when that outcome is not only foreseeable but desired as an end, and contrast this with side effects that are foreseeable but arise as by-products of the pursuit of some other end. Maybe they’ll be consistent in applying this notion of intentionality, or maybe they’ll deploy it differently in different contexts. The only way to find out is to ask.

      Now, I sense that the big worry is that because the concept of intentional action can be formulated in more or less sophisticated ways, then asking people who haven’t explicitly thought about the concept is pointless: it might not tap into anything deep. First, I think this worry is oversold as people don’t just talk about intentional behaviour willy nilly – it often comports with what a reflective account of intentional action might be. That is, ordinary people do use the concept of intentional action to describe someone doing X in order to achieve Y, and distinguish this from doing X in order to achieve Y, while producing Z accidentally. But then there are these funny quirks like the Knobe effect that suggest that ordinary people, in some circumstances, apply the concept of intentionality in way that trained philosophers might not. But if we want to understand the inner workings of the social mind, it’s probably best to look at how ordinary people think rather than the tiny percentage of humans globally who indulge in deep, analytical philosophical reflection about these matters.

      Second, you could say the raise the same concerns about statistics, which is a deep, often confusing subject. Now, mathematicians may have very sophisticated conceptions of statistical notions compared to most other people. In fact, humans’ statistical judgements often depart from what maths tells us (at least when problems are framed in certain ways). But it’s nonetheless interesting to see how people go about statistical reasoning, even when they don’t have a sophisticated understanding of statistics as practiced by the experts. Indeed, it’s because they’re not experts that we’re interested in how they reason in this domain: we want to know how people — people in general, not an elite group of brainiacs – reason mathematically. Just as studying ordinary people’s mathematical intuitions may not help you solve mathematical puzzles and theorems, so too studying folk judgments of intentionality may not help resolve debates about what should count as intentional action. (Though such studies may be relevant to philosophical disputes in other ways – too much to get into here for the moment.) Nonetheless, they provide a starting point for thinking about how people naturally, intuitively make sense of the social world and others peoples’ behaviour.

      I don’t feel I’ve done this reply justice. The X-phi folks should have more to say!

  2. David Rose says:

    Thanks so much for the post. I wanted to take the opportunity to mention a recent paper, forthcoming in Philosophical Psychology, that my co-authors, Jonathan Livengood, Justin Sytsma and Edouard Machery, and I, wrote in response to the work of Sripada and Konrath. In short, we raise a series of technical objections in response to the modeling work done by Sripada and
    Konrath, showing that their own data undermine the Deep Self Concordance Account as it is currently articulated. Our paper, as well as a discussion of it can be found here:

    And for those interested, there is an earlier discussion of our paper here:


    • Dan Jones says:

      David, thanks for the further pointers. I had a look at the posts/paper/debates, and I’ve got to admit that like Donny in the Big Lebowski, I’m out of my element here. This is a pretty hardcore statistical/methodological dispute, and I don’t pretend to be able to adjudicate on it. To everyone else, roll up your sleeves and get stuck in if you fancy it! (I’ll take a closer look at the paper, and see if I can summarise some of the issues at play.)

      • David Rose says:

        Hi Dan,

        Maybe I can help a bit with the summary of the paper. As you note in your post, Sripada and Konrath use SEM to model the causal relationships between the various variables that relate to the DSCA as well as the normative models of interest. Using this technique, they show that the DSCA variables but not the normative model variables
        causally influence intentionality judgments. Our first objection is aimed at the specific modeling technique that they used. Essentially, we suggest that standard use of SEM is akin to “guessing and checking” whether one’s favored model fits the data. The issue with this, as we point out, is that there are many causal models that might fit the data and these models might be incompatible with one’s preferred model. So, we used a more principled approach to causal modeling; that is, we use a search algorithm which is designed to search over the data and find the best fitting causal model of the data. When we do this, we find a model that fits the data better than Sripada and Konrath’s, but is incompatible with the DSCA.

        For our second objection, we begin by pointing out that Sripada and Konrath built an SEM that included both DSCA variables and the normative model variables, concluding that only the DSCA variables causally influence intentionality judgments. But one can ask why this overall model fits the data. That is, does it fit the data well because the variables that encompass the DSCA are especially good predictors of intentionality judgments or does it fit it well because the other variables in the model fit the data particularly well? What we show is that, when one looks at the variables that encompass the DSCA that these variables fit the data very poorly (poorly enough that the DSCA is rejected by a chi-square test), but that the variables for the normative models fit the data very well. Thus, the overall fit of the model is due to the excellent fit of the variables associated with the normative models, variables that do not predict intentionality judgments. Thus, neither the DSCA nor the normative models end up explaining intentionality judgments.

        Our final objection is aimed investigating whether Sripada and Konrath’s model satisfies certain causal modeling assumptions. Roughly, the Markov and Faithfulness conditions ensure that certain independence and dependence relations hold between variables in the data and allow for a causal interpretation of the data. What we show here is that the DSCA violates these assumptions and so the causal relationships between the DSCA variables and the intentionality variable that Sripada and Konrath model cannot be correct. From all of this, we conclude that the DSCA, as articulated, cannot be correct. Interestingly, not only does the version of the DSCA tested fail to explain intentionality judgments, but the normative models tested also fail to explain those judgments.

        • Chandra Sekhar Sripada says:

          I really hate these public exchanges where you make overheated claims and I am forced to respond. But what am I to do, ignore these claims? So here we go again… I am sorry to everyone out there in the blogosphere about the harsh tone of this reply. I just don’t know what else to do in defending the study by Sara and me.

          As many know, I *strongly* disagree with the Rose et al claims. People need to know that the methods used by Sara and I are very standard (more than 10,000 studies according to PsychInfo). So we represent the received view and Rose et al (who do not have a single paper on structural equation modeling) represent a dissenting voice. The posts from the XPhi blog contain the key elements of my responses to them. A formal reply to their piece, co-authored with leading figures in statistical learning theory, contains the details. Let me summarize the key points.

          Rose et al make three arguments. Argument 1 is based on a statistically illegitimate analysis – you can’t search with a scoring variable and use that very variable to measure the goodness of fit of a model. This is stats 101, and is *unequivocally* an overfitting error. I am surprised that they haven’t revised their paper yet. I will definitely take this up with the editors of the journal that accepted their piece.

          Arguments 2 and 3 are best addressed with a simple story. Suppose a researcher proposes a “HeadStart model” that claims that a HeadStart intervention in pre-school will decrease use of drugs and days of school missed, and will thereby enhance academic outcomes. Sure enough she finds that receiving the HeadStart treatment significantly predicts less use of drugs and days of school missed and these variables in turn significantly predict better academic outcomes. The data support the model right? Not according to Rose and colleagues. They come up and tell the woman that there is a significant causal effect between using less drugs and missing less days of school, and this effect was not in the original model. Therefore the model, “as it is currently articulated”, is falsified. This of course sounds ridiculous. The woman may not have put that link in the original model, but it is 100% in the spirit of the model that such a link exists. To say that her model is thereby falsified is absurd to the point of making people laugh. When you unpack the opaque and unnecessary statistical language from the Rose et al piece, arguments 2&3 amount to just this ridiculous claim (notice the repeated use of this “as it is currently articulated” qualifier throughout their posts and paper).

          My co-authors on my reply to the Rose et al critique include Richard Gonzalez and Vijay Nair, both experts in statistical learning theory (and former chairs of their departments at Michigan). Am I making an argument from authority? Hell ya!! Regular people can’t unpack the dense statistical language in the Rose et al piece and understand my responses. So I am happy to appeal to authoritative experts who conclude their claims are way way off base.

          On a final note, I and my colleagues did a bunch of follow-up studies. We have now studied more than 2000 subjects. The model proposed by Rose and colleagues is falsified repeatedly by the data. The best way to settle a substantive dispute is to let the data talk, and it has, RESOUNDINGLY. Again, I apologize to everyone out there for the tone of this response. Methodological disputes can be helpful but I am afraid this one generates more heat than light, and I am frustrated at having to rehash my dispute with the Rose et al claims yet again.

          • David Rose says:

            Hi Chandra,

            Thanks for your post. I do not want to rehearse this dispute here, but I was hoping that interested readers could get to see both sides of this issue. So, for readers who are interested in this exchange, Chandra makes somewhat similar points here:
            and my co-authors and I provide responses. In short, we disagree with Chandra’s responses and suggest that they miss the point.

            All the best,


          • Justin Sytsma says:

            Dear Chandra,

            I am sorry that our attempt to engage with your work, and the interesting methodological questions that it raises, has you so worked up that polite exchange is impossible. Lesson learned. For my part, it won’t happen again, as I have no intention of engaging with your work further.

            Wishing you all the best,

            • Dan Jones says:

              Hi Chandra, David and Justin,

              Thanks for visiting and leaving comments, though I appreciate it might not have been entirely pelasurable. I’m not going to insert myself into this dispute, but I felt I should acknowledge that something’s going on. Obviously this thread has history over at the Experimental Philosophy blog (which I wasn’t previously aware of), and as I’m at sea with this particular set of issues I’m going to have to leave this argument here. I already feel like a bit of an intellectual arsonist, reigniting a fire that had died down, and I think I’d only add confusion to the mix. I hope everything can be smoothed out.

              All the best,


  3. People are held responsible for thoughtlessness that results in a bad outcome while commonly not given credit for thoughtlessness that results in a good one.

    “These results — and they’ve been replicated many times since — are a bit surprising.”
    They’re not surprising at all, they’re obvious. The fact that so many academics think otherwise is a bit shocking.

    “Intention” is not the best -most appropriate/most clear- word to use to capture the motivations for people’s response. But it’s the word you, Knobe et al., choose, for your own reasons. Subjects used the words given them to describe their reasons only as best they could.

    People have an interest in promoting socially productive behavior. Thoughtlessness is thoughtlessness. It’s not something we should encourage. Put this way “the Knobe effect” is a truism, and old as the hills.

    • Dan Jones says:

      Hi Seth, thanks for dropping by. You suggest that the Knobe effect is a bit old-hat, pretty obvious stuff. Maybe so. But it still has to be explained. Saying “People have an interest in promoting socially productive behavior. Thoughtlessness is thoughtlessness. It’s not something we should encourage” doesn’t cut it as an explanation – though I think it does point in the direction of an explanation. That is, the asymmetry in intentionality judgments might exist so as to underpin the punishment of bad behaviour, while not rewarding behaviour that just happens to achieve something good. (This could make sense in evolutionary terms, for enforcing groups norms of behaviour and contributing to cultural group selection.) Or it could be a by-product of the way the way the mind is hooked up. These are interesting questions.

      I had a look at your blog, and this sentence stood out: “The really idiotic, morally sleazy aspect of the focus on intention is that it focuses on the desires of the actor, his sense of his own emotional state, his “sincerity”, rather than the outcome.”. I’m not quite sure who this is directed at: researchers investigating this stuff, or regular folk who focus on intentions in their moral judgments? If you find this research idiotic or sleazy, we’re not going to find much common ground here. And you say, in response to my suggestion that we might assume that judgments about intentions feed into moral judgments about actions and outcomes, that this is “liberal self-regard”. I don’t know what you mean, but it doesn’t sound good! (I am a liberal, but I don’t consider what I wrote to be particularly self-regarding.) Still, I welcome your comments here — perhaps you’re speaking for the majority.

  4. “There’s a world of difference between me walking up to someone in the street and punching them square in the face, and accidentally hitting someone in the face as I’m trying to provide directions to another passerby (a feat I once achieved on London’s Tower Bridge). In the first case, my intentions would be bad, and so the outcome is judged as a moral offence; the latter was an accident, a side-effect of trying to help someone else, and therefore not morally blameworthy.”
    Simply not true. If that were the case there would be no charge on the books for manslaughter

    “Between 1947 and 1949 an odd group of idealists and hard realists in the American government set out to intervene in Syria. Their aim was to liberate the Syrian people from a corrupt autocratic elite – and allow true democracy to flourish. They did this because they were convinced that “the Syrian people are naturally democratic” and that all that was neccessary was to get rid of the elites – and a new world of “peace and progress” would inevitably emerge.

    What resulted was a disaster, and the consequences of that disaster then led, through a weird series of bloody twists and turns, to the rise to power of the Assad family and the widescale repression in Syria today.”
    The full story is here

  5. Dan Jones says:

    You don’t deny that there is an important distinction between me punching someone in the face with malice and catching someone in the face as I move my arm to point someone else in the direction they need to go. But you do imply there’s no moral difference – or at least that my accidental smack is not morally neutral. Really? Do you not distinguish, morally, between acts of deliberate violence and accidents? And yes, you can non-deliberately cause harm to someone and be held morally and legally responsible, as in manslaughter. Yet there’s a legal and moral distinction between bona fide accidents, negligence, manslaughter and murder. And these carry different moral weights – don’t they? By the way, none of what I’m saying is a prelude to justifying foreign policy decisions by suggesting that whatever harms they caused were unintentional and merely the by-product of some other supposedly moral goal. (I only add this as it seems to be the implication from the allusion to Syria, though I may have misunderstood the supposed relevance.)

  6. Tomkow says:


    Thank you for replying and thank you for the kind words about my blog.

    You respond to criticism by rehearsing an Xphi manifesto. This is standard practice for Xphi proponents. But I wasn’t complaining in principle about the Xphi program. I was saying it was bad experiment.

    If you ask the man on the Clapham Omnibus if what Knobe’s agents did was right or wrong he will answer without hesitation. But if you ask him if the actions were “intentional” he will not answer ‘yes’ or ‘no’ he will ask, “what do you mean”?

    Which would be a good question because “intentional” can mean many different things . It can be read as synonymous with: ‘deliberate’, ‘voluntary’, ‘premeditated’, ‘purposeful’, ‘unforced’, ‘un-coerced’, ‘witting’, ‘willful’, ‘motivational’, ‘designed’,’ planned’, ‘studied’, ‘not-accidental’, ‘advertant’, ‘conscious’ … Each of these means something different and we know these differences are ones that folks are sensitive to precisely because these different terms survive in folk parole.

    Which of these senses of ‘intentional’ did Knobe have in mind when he asked his questions? How can we sure that his subjects were reading the term in the way he intended or the same way as each other? How can we be sure that they were interpreting the term in the same way in the same ratios in the different scenarios presented? How can we be sure that we are measuring the role of moral factors on judgments of intentionality as opposed to the influence of moral factors in the disambiguation of ‘intentional’.

    Absent controls for these variables and their permutations the results are meaningless.
    You might say these are all matters to be sorted out with more experiments, but before we can figure out what further questions to ask we need to be clear what we are trying to find out .

    Knobe’s subjects didn’t get to ask “what do you mean?” We can.

    The question I would put to him— and all the other participants in this research program– is what *they* think the *correct* answer to Knobe’s questions are? Why? And why does the answer matter?

    To answer those questions they will , I’m afraid, have to do some non-experimental philosophy. But If they cannot answer these questions for themselves they have no business pestering other people with questionnaires.

    • Dan Jones says:

      Tomkow, thanks for the further comments, and apologies for the slow reply – been a bit preoccupied with some other things. You suggest that the man on the Clapham omnibus could easily answer questions about the rightness and wrongness of an agent’s actions, but would ask “What do you mean?” if asked about whether an agent achieved some outcome intentionally. How do you know? Do you have survey data to support this? I’ve run the Knobe scenarios by many people, and no one’s ever asked me, “What do you mean?”. They answered quite happily (I concede that my friend’s might not be representative of omnibus passengers!).

      You also suggest that asking about intentionality is inherently problematic because it could mean one of many things (‘deliberate’, ‘voluntary’, ‘premeditated’, ‘purposeful’, ‘unforced’, ‘un-coerced’, ‘witting’, ‘willful’, ‘motivational’, ‘designed’,’ planned’, ‘studied’, ‘not-accidental’, ‘advertant’, ‘conscious’). But couldn’t the same be said of asking about right and wrong? On the same kind of analysis as you applied to ‘intentionally’, right and wrong could be taken to mean ‘good’ or ‘bad’, ‘(im)moral’,’sinful’, ‘decent’, ‘criminal’, ‘appropriate’, ‘illegal’, ‘evil’, ‘tasteless’, ‘(im)permissible’, ‘proscribed’, ‘taboo’, ‘blameworthy’, ‘praiseworthy’ etc etc. Does this mean that we can learn nothing by asking regular, non-philosophers (and non-dictionary compilers!) whether they think something is right or wrong? As you say, “Each of these means something different and we know these differences are ones that folks are sensitive to precisely because these different terms survive in folk parole.”

      Now, the fact that there’s a systematic difference between the responses elicited by Knobe’s chairman cases seems to point to something interesting. I suppose it could be the case that in both conditions, people have no idea what on earth is meant by intentionality, and settle at random on one of the various meanings you provided – but if so, why is there is such a robust, systematic difference in responses across cases? Or you might suppose that the two different conditions activate different intentionality-type judgments – perhaps the harm cases activate concepts of intentionality that relate to foreseeable consequences, whereas the help case activates concepts to do with actually desiring the outcome. To put it another way, perhaps respondents would say, “When you asked me whether the Chairman intentionally harmed the environment, I took that to mean ‘Did he know that the environment would be harmed?’. But when you asked me whether he intentionally helped the environment I took that to mean ‘Did he desire to help the environment as a way to make a profit?”. If this were the case, there’d be no strict contradiction or inconsistency – the different answers are responses to two questions that are perceived to be different. If this is the case, that’s still interesting – why do the two cases activate these different concepts of intentional action?

      In any case, this possibility has been explored by asking whether the ‘Chairman harmed/helped the environment in order to make a profit’, rather than asking about intentionality per se. In these cases, the concept in play has been restricted to a more specific meaning, namely, that the Chairman does X as a means to achieving Y, rather than X arising as a side effect of achieving Y. As before, people tend to say that the Chairman did harm the environment in order to make a profit, but did not help the environment in order to make a profit. This again suggests that the two cases cause people to deploy the same concept (‘intentionally’ or ‘in order’) differently in the two cases. I do not see these as meaningless results, and this asymmetry requires explanation, rather than dismissal.

      Finally, it’s not clear to me why the people who carry out this research have to articulate a notion of intentional action. Does someone who studies what actions the general population considers to be moral or immoral have to have a fully worked out moral theory that adjudicates between the cases they present to their subjects? If so, why? Likewise, couldn’t I be agnostic on the problem of free will, yet still ask other people what they think about it, and get meaningful answers? (I appreciate that if I’m completely muddled on the notion of morality or free will, I’ll probably make a mess of asking people about their views!)

      But let’s say I did have a concept of intentional action, one that means the following: “The outcome of an agent’s acts are produced intentionally if she foresees the outcome, desires the outcome, and is appropriately causal connected to the outcomes”. (I know, this definition raises its own set of problems). And then say I went out and presented people with the Chairman cases, and got my answers. What difference would it make if I had different idea of what counted as an intentional action? After all, we’re interested in what people per se think about intentional action, not what I or professional philosophers do.

      I’m sure we’re no closer to agreement on the significance and importance of the Knobe effect, but thanks for the stimulating exchange!

  7. Neil Rickert says:

    Thanks. An interesting post.

    Firstly, I don’t think the situations (“harm” vs. “help”) are symmetrical, so I don’t find the results particularly surprising. Before I explain that some more, let also add that I would find it very hard to judge the first (“harm”) case, and far easier to judge the second. And the reason is that there is too little information. Was this to cause minor, barely significant harm to the environment? Or was this likely to be a flagrant violation of environmental laws, or was it somewhere in between. It seems to me that we would be less likely to ascribe intentions if it is a minor harm than if it is a major harm.

    As for the non-symmetry: we are part of a culture, and that culture assumes that people have responsibilities toward one another. Specifically, we assume that people have responsibilities not to harm others. That’s what tort law is all about. That’s why we carry liability insurance on our automobiles. However, we do not have corresponding assumptions that require people to do good to others – we only assume that they will have a responsibility to not do harm. And that difference is what makes for an asymmetry between the cases.

    • Dan Jones says:

      (I thought I posted this yesterday – my withered brain was mistaken, so here’s what I had to say, again)

      Hi Neil, thanks for leaving a comment, and the prompt to further reflection on this fascinating topic. I think I get what you’re saying, but I don’t think it can explain the Knobe effect. Yes, we have special responsibilities to not harm others in a way that doesn’t apply to helping them (though in some countries not assisting someone in need is considered a crime in some circumstances). But a couple of points. First, the Chairman cases ask about intentionality, not responsibility – and you can be responsible for things that you don’t do intentionally (if I accidentally leave a lit cigarette in someone’s house, causing it to burn down, I’m responsible even if I didn’t burn the place down intentionally). Second, if failure to live up to the responsibility to not harm the environment is the key factor here, why can’t people just say that in both cases the impact on the environment was unintentional? It would still be perfectly sensible to say that he was nonetheless responsible for harming the environment and simultaneously guilty of not living up to the maxim of to do no harm (or at least let no harm happen on his watch, and therefore deserving of moral censure. But on this account, isn’t the Chairman equally responsible for helping the environment? And if we don’t have special responsibilities to do good, then doesn’t he deserve special praise for this, for going beyond his obligations? No, because he didn’t really care about helping the environment – it was just a lucky accident, a side effect of pursuing the profit motive. It was unintentional, as the majority of people agree. Yet the same logic applies to the harm case, and here the majority of people the bad outcome was intentional – even though this judgment isn’t necessary to support the claim that the Chairman failed to live up to his responsibility to not harm the environment, and is therefore deserving of moral blame. It does seem to me that the concept of intentionality is being deployed in one case (harm) and not the other (help), even though there is a logical symmetry between the cases. Nevertheless, this asymmetry in responses must be due to an asymmetry in other factors: perhaps normative, or, according to the DSCM, concordance between attitudes and outcomes of actions (or perhaps something else entirely). But in short, I don’t think that responsibility is the key issue here. (I’m not convinced that this is the clearest explanation, but excuse me – long week, bit frazzled.)

  8. Rich Boulton says:

    Really interesting stuff, thanks.

  9. I want, or need, to add something else, since my response was too directed at the origin of this discussion in, Knobe’s claims. I haven’t changed my mind about them, but this paper deserves credit for making something interesting out of what I still think of as banal. What it shows- and I’m relying on the synopsis here- is not how decisions seemingly based on normative belief are based on a search for consistency, but how much we to we mask our moral assumptions in logic. This ties into my response above to claims about what is and is not “morally blameworthy” but its clear that Sripada and Konrath have given us new information about why Dan Jones would make that claim, and why they make the claims they do about data that shows them to be wrong.

    “The basic idea is that when we encounter cases like Knobe’s Chairman scenario, we make two different judgments: one about his attitudes and values — is he anti- or pro-environmental? — and another about whether the outcomes of his actions are consistent with those attitudes. Neither of these judgments is normative; they are descriptive judgments about the facts of the case.”

    But “logically” the response would be to take the chairman at his word and say that he was neither pro or anti environmental. According to his statements, absent perceived moral implication, the question would be the equivalent of whether or not a pebble would end up on his desk or a glass would contain water or seltzer. The responses documented in the paper are based on assumed and implied values, but only after being shrouded in logic; shrouded to a degree that the authors, like the subjects, like the author of this post, saw the logic but not the moral foundation. This, with the discussion of the ventromedial prefrontal cortex, makes this paper much more interesting than anything by Knobe.

  10. Pingback: Atoms in the Void (1 July 2011) « The Productive Philosopher

  11. Florian Cova says:

    Thanks for this great and precise summary of Sripada’s solution to Knobe’s problem. I will shamelessly profit of this occasion to let know that I have a forthcoming paper on this topic, where I argue that there are cases that Sripada’s model cannot account and that the model should be “corrected” to make place for the influence of normative factors. You can have a look at it here :

  12. Mike says:

    I want to preface this post by saying that I do not intend to make a criticism—I intend for this to be a question (which hopefully hasn’t already been addressed elsewhere). Here it goes:

    It seems to me that when a “price” is paid for an apparently greater profit, then the one who makes the decision to receive the profit is responsible for the price that is paid (the means that result in an end). However, when an unrelated good is a consequent of receiving a profit then the one who receives the profit is not credited with the authorship of the good in question (since s/he is aiming for the end—profit in this case).

    For example, when we see, “I don’t care at all about harming the environment. I just want to make as much profit as I can” in the section above, what the reader takes from this is that the chairman is willing to sacrifice the environment in exchange for making more money. Conversely, when we see, “I don’t care at all about helping the environment. I just want to make as much profit as I can” we infer that the benefit to the environment is not of interest to the chairman. The only thing s/he is interested in is profit. It follows from this latter statement that the chairman would have started the project even if there was no benefit to the environment and thus the benefit of the environment was not his intention (rather the profit was)—as opposed to the harm-case, in which harming the environment was a means to an end and thus was intentional.

    I’m not sure that I have sufficiently articulated my reasons for believing so, but isn’t deducing that “the ‘moral flavour’ of the outcomes of actions — that is, whether they’re good or bad — shapes whether or not we construe those actions as having been performed intentionally” sort of begging the question (since in the harm-case an intentional decision to cause harm to the environment as a means to an end is implied, but the help-case the profit is stated as the end and the helping of the environment a result which the chairman cares nothing about)?

    • Mike says:

      Perhaps eventually someone will respond to this?

      • Dan Jones says:

        Sorry Mike, totally went off radar, my bad. I’m not sure I fully get your point, however. You write, “It follows from [saying “I don’t care at all about helping the environment. I just want to make as much profit as I can”] that the chairman would have started the project even if there was no benefit to the environment and thus the benefit of the environment was not his intention (rather the profit was)—as opposed to the harm-case, in which harming the environment was a means to an end and thus was intentional.”

        That’s not my reading, which is that in both cases the Chairman would’ve started the program irrespective of its environmental impact. That is, just as the Chairman would’ve started the program even if there was no environmental benefit, he likewise would’ve started the program had it not caused environmental damage. So in the both cases, the harm/help are equally side effects of the profit motive, and so I don’t see why one should be seen as a means to an end and the other not. Yet still, they feel different! And we’re back to square one 🙂

  13. Luke Gibson says:

    It seems to me that we judge negative outcomes as intentional when they are a result of carelessness, and thus a kind of intentional irresponsibility.

    The harm case constitutes carelessness on the part of the chairman and so we deem it intentional, making him accountable for his irresponsibility.

    We resist considering any positive effects of carelessness as intentional because it would undermine our low opinion of carelessness.

    I cannot see how the Knobe effect is doing anything other than illuminating our dislike of carelessness and our conviction that it generally results in negative outcomes. Both models above seem to miss this.

    Now, perhaps this attitude towards carelessness is an unwarranted bias… But to challenge our perception of carelessness we would need evidence to suggest that carelessness generally results in good, or at least not bad. We would need to look at the net results of carelessness, which is easier said than done.

  14. Pingback: linkfest – 07/10/11 « hbd* chick

  15. Ian Shipman says:

    I have a naive question. (I do not have time at the moment to read through all the comments, so apologies if this was asked & answered already.) In figure 1, there is an arrow directly from cases to intentionality judgments. It sounds like this is meant to stand for the part of the effect that is not soundly attributable to deep-self variables. Does this mean that the DSCA explanation is incomplete? If so, are there any ideas about what other features on the mental landscape might be responsible for this?

    • Dan Jones says:

      Yes, that’s right, the arrow from Case to Intentionality signifies effects that are not mediated by deep-self variables, so these can’t be a complete explanation (but on this analysis at least the normative variables are not even a significant part of the explanation of the Knobe effect).

  16. Gatogreensleeves says:

    Great post!
    I have a silly question: am I to understand that the descriptive data in DSCA operates like a cum hoc ergo propter hoc, wrapped in bows of confabulation and availability bias? Is this oversimplifying, considering the normative influence redirection in Fig. #2?

  17. Pingback: The Anatomy of Intentional Action (意向性行動的解剖學) « 法哲學、生活與實踐

  18. Pingback: DPM meeting: Should philosophy be experimental? From 5-6.30pm in ib3.307, WP campus » Deakin Philosophical Society

Have your say!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s