Blog Coordinator

« Neuroscience and Free Will Revisited | Main | Recent Work on Agency and Responsibility »



Feed You can follow this conversation by subscribing to the comment feed for this post.

I don't think that toxo, and the like, changes anything - not notions of the self and not free will. Not the self because unity has always been a fragile and unstable achievement. We unify ourselves, limiting the degree of conflict between the modules which constitute our minds. Not free will, because toxo is simply another influence on our dispositions, akin to our biases and heuristics. It just has an extra neural realisation, or perhaps better, an extra axobic realisation. These are not alien to ourselves, but partially constitutive of who we are. Thinking otherwise is just a version of the internalist prejudice. *That's* not the reason to be a skeptic.

Well, if toxo made humans as reckless as it makes rats, that would be a pretty serious dent in our reasons-responsiveness.

Right, Paul. But since we are not in general reckless, this would seem to suggest that our executive faculties are not significantly compromised by such organisms. Our decision-making, which brings to bear our capacities for reasons-responsiveness, certainly takes place against a background of "givens"--factors (as Neil nicely points out above), including biological factors, that incline us one way or another.

Hmmm...I am not sure about the responses by Neil and John (and I am with Paul)! Let me frame my reservations by looking at mice and not men for a moment. Take two mice: One infected by toxo, the other not infected by toxo. The former is now prone not only to impulsive behavior, but impulsive behavior that runs counter to its best interests and even survival (e.g., being attracted--rather than repelled--by the smell of cat urine). It seems clear as day to me, at least, that the agency and conscious control of the infected mouse has been compromised (especially in relation to the non-infected mouse).

Why, then, do we not say the same sort of thing in our own case? Let's say the estimates are correct and between 30-50% of people are infected (and hence affected). If we assume that the manipulation hypothesis is correct and we assume that whatever micro-manipulations caused in humans happen below the veil (and often encourage impulsive actions that are against our interests), why not think the resultant behavior is less free and responsible than the behavior of someone who is not infected?

At a bare minimum, it would seem that the infected have more impulsive inclinations and desires to resist than their non-infected counter-parts--which, in itself, seems to affect how much agency, control, and blame I am inclined to attribute to them (here again, relative to the non-infected).

That's partly why pointing out the fact that we're not reckless "in general," as John does in his comment above, doesn't adequately address the problem at hand. The question is not whether we are reckless in general--since this issue averages over the behavior of the infected and non-infected. The question is whether the infected are sufficiently reckless in relation to the non-infected such that we should say the agency of the former has been (partly) compromised.

Here again, I think we most certainly would say this in the case of mice (and they, too, I would argue, have reasons-responsive mechanisms). So, I don't see why we shouldn't say it in our own case as well.


18-22 yr olds often engage in impulsive behavior that is contrary to their own best interests. Perhaps, then, their self-control is compromised relative to what it could be if they weren't so reckless.

It seems to me that generalization will always be true: the reckless among us will be less in control (in risk-assessing ways) then those of us who are not particularly reckless. This isn't enough to say that they aren't enough in control to be responsible, or that in all things we apportion proportional to degrees of control. Maybe we should say that, but such a claim seems to me controversial.

It is true that in the case of the parasite, there is an extant explanation for just *why* those with it are reckless. And maybe the hope is that for every instance of recklessness, we'll be able to pick out the very neurological correlate for it. Does it matter whether it is a structural correlate or a microbial one?

I'm inclined to think not, anymore than there's a correlate for a loss of control (in that I tend toward acting against my best interests) in myself when my blood sugar drops. But maybe there's an argument to be made that its being caused by a microbe does make a difference. But then that claim would need to be explicitly defended.

Matt, You're making the same kind of move John was making. I already tried to point out why it's a mistake to generalize across all humans.

Your contrast between 18-22 year olds and older adults is similarly problematic. The right contrast would be a within-group comparison of the 18-22 year olds.

More specifically, we would compare the 30-50% of these individuals who are infected by toxo with those who are not. The question I was raising is based on the assumption that the toxo infection confers an added risk for impulsivity (above and beyond age). If that were true in this case, then why wouldn't we say that the infected 18-22 year olds are less in control of (and hence less responsible for) their impulsivity than the non-infected?

As for your other analogy--namely, low blood sugar--it seems inapt. My blood sugar isn't manipulating my neuro-architecture and behavior in order to increase its chances of reproduction. It also hasn't been linked to mental disorders such as schizophrenia, etc. Finally, my blood sugar isn't alien to me in a way that toxo is alien. Toxo is a parasite--a micro-parasite that may have the ability to manipulate and influence my macro-behavior without my awareness. That strikes me as importantly different than the other kinds of structural causes you seem to have in mind.

That said, I feel like we aren't doing justice to the issues raised in the original article (which is partly my fault). I have been focusing on toxo because it's the piece of Kramer and Bressen's puzzle with which I am the most familiar. They present a multi-pronged attack on the notion of a unified self. Toxo is just one small piece--albeit one I find especially interesting.


Good, and thanks very much for pressing me. I don't know the details of the research. But here's what I'm wondering. I suppose that due to factors quite beyond our control, we are dealt a certain hand, as it were, by our genetics and early childhood experiences. These factors make some people much more inclined toward anger, fear, anxiety, and so forth, than others. And yet we hold them morally responsible, or typically do. We think moral responsibility involves decision-making (perhaps weighing reasons and exercising one's capacity for reasons-responsiveness) against a backdrop of proclivities and inclinations that are just "given exogenously", as it were. Why are the bugs, and their effects on us, any different? Why don't they just set up certain predispositions and inclinations, against which we can still display the signature and distinctive human capacity for reasons-responsiveness?

Life is tough, and unfair. The natural and social lotteries (to borrow the metaphor from Rawls) are brutal, and leave people with differential propensities toward anxiety, fear, depression, and worse. No doubt, illness and subacute biological maladies also affect our dispositioins. But why wolud they thereby eliminate or etiolate our moral responsibility, and more than genetics and early childhood experiences?

Now maybe you, as a moral responsibility skeptic, will say, "Right--there is no difference, and they all rule out moral responsibility." My inclination is to go the other way. It would be interesting to figure out how to resolve this sort of dispute.

I can't help but notice that the framing of the issues in the abstract quoted in the post sets things up in, what seems to me to be, a biased manner. [I'll admit up front that I haven't read the entire paper.] Consider, for example, the authors' summary of what they are up to: "This article provides a broad overview ... of the consequences of our coexistence with these selfish entities." This way of putting things frames the issue in terms of a struggle between distinct selves--us, on the one hand, and various microbes, genes, DNA and brain cells, on the other.

When a microbe, considered as external to me, is found to be influencing my behavior, the natural implication is that this influence is external to my agency. But, it seems to me, this is to beg the question. As Neil's comment demonstrates, there are real issues (normative issues!) at the heart of the matter. What counts as a unified self? How unified must a self be in order to act (freely)? When does the influence of some factor count as alien with respect to my agency? The way this paper frames things, however, the answers to these questions are all built in at the ground floor. The conclusion that we are super-organisms with no free will simply follows from the description of the findings. But why accept this way of putting things?

Here is an alternative description, a paraphrase of what the authors say they are up to: This article provides a broad overview of the many surprising factors that influence our behavior from inside our skulls. This way of putting things leaves it entirely open whether the factors the authors go on to discuss, which are, undeniably, *internal* in a significant sense, are alien to our agency or not.

As I see it, the determination that, say, toxo is an alien force influencing one's behavior is a normative matter. I can agree with Thomas that it is useful to consider the contrast between those who are under the influence ["infected" is a normatively loaded term] of toxo and those who are not. But the fact that those under the influence of toxo behave differently from those not under the influence of toxo does not settle the central issue, namely, whether this matters. We are all built differently and we are finding out in more detail every day about the various influences on our behavior. But the mere fact that my behavior is subject to various influences does not yet show that it is subject to alien influences. That further claim relies on normative claims about what is and is not internal to my self, in the relevant sense.


Toxoplasma gondii is a parasite (a very clever one that happens to have the fairly rare power to break the blood-brain barrier). It can lead to toxoplasmosis--which is an illness that can be life threatening in those with weakened immune systems. It also appears to have the ability to influence its hosts' behavior in subtle and not so subtle ways. Yet, you say the term "infected" is normatively loaded. Another term we use with parasites is infested. Would you prefer that to infected? Either way, given the very nature of the parasite-host relationship, it would seem that parasites are alien in some important sense. How else would you explain the relation in the case of toxo? What non-normatively-loaded term would you use to capture the relationship? This is one sense in which toxo is different than some of the other micro-organisms discussed by Kramer and Bremmer. There is a difference between exogenous and parasitic brain microbes like toxo and exogenous viruses like cytomegalovirus and the "good" bacteria in our gut--namely, the former confer risks while the later confer benefits. But there are other differences as well--namely, the fact that toxo and cytomegalovirus are trying to control or even attack us.


Thanks for your reply. It seems that we agree on the main point of my original comment. The claim that a given influence on one's behavior is alien rests, in part, on a normative foundation. For example, as you point out, there is a difference between various kinds of exogenous organisms that enter into our bodies and influence our behavior, and this difference is based on normative considerations, such as whether their influence on our behavior is beneficial to us or not.

But you seem taken aback by my comment that "infected" is a normatively loaded term. All I meant was that we apply this term to exogenous influences that we take to be bad. So I would have thought you'd agree that the term is normatively loaded. And I fully agree that the parasite-host relationship is to be characterized in normative terms. The term "parasite" is itself a normatively loaded term. We don't call organisms that help their hosts achieve their (i.e., the hosts') ends parasites. So, again, I think we are on the same page here.

I don't even really have a bone to pick with your claim that toxo is a parasite. Perhaps it is. All I wanted to point out is that one cannot fairly set up the discussion of whether its influence on our behavior is alien by describing it is a parasite--or even in terms that distinguish between its ends and our ends. The very description assumes the conclusion one is trying to establish. And from the abstract you posted, it seemed as if the authors of the target paper did just that.

I think this is worth pointing here because it is an easy trap to fall into. And it makes it seem as if the conclusions that we are super-organisms and lack free will are obvious given the empirical findings. But these conclusions need to be established by argument that appeals to normative premises; they cannot be established simply by pointing to the fact that there are, as a matter of fact, various detectable influences on our behavior.

Another easy trap is the one that says that toxo (or any other micro) "controls" or "manipulates". There are perfectly good uses of these terms on which those claims are true, but not the rhetorical ones meant to threaten free will, responsibility and our sense of self. I don't think anything can manipulate me in the threatening sense without mental states, and I'd be hard pressed to think toxo has those. It certainly doesn't have the requisite intentions to count as a manipulator.


I don't think I was making the same move as John, since I wasn't trying to generalize across the entire population. Instead, I was just comparing the reckless to the less reckless to note that, by that comparison, the reckless will always seem to have their agency somewhat compromised -- at least in the sense that they make self-harming choices. But that's equivalent between adolescents and those infected with toxo.

I understand that you take the fact that it's an exogenous microorganism to be conclusive for showing that agency is compromised in the toxo case but not with other structural factors. I'm inclined to disagree.

Now, I don't expect the initial burden to fall one way or the other on this disagreement. But insisting that the only comparison to consider is between those with toxo and those without is to exclude comparisons that may be relevant, at least without first ruling them out. But then the burden does fall on you to say why toxo is different, since the point of the comparisons is to tell us something about whether toxo might compromise agency.

It seems to me that this discussion is another form of the classic "is/ought" issue. Namely the claim that "X is a superorganism" entails that "X is only arguably responsible for X's actions." So, a move from a constitutive/identity "is" to a normative "is", which probably runs roughshod over several other "is-s" that presumably connect them as logical claims embedded in connecting arguments.

I just saw Ben's comments. So now I just add this as a call for clarity in making these connections. I have no dog in this race.


You’re absolutely right when it comes to how I would respond to your “school of hard knocks” approach to the lottery of life, etc.! Indeed, it seemed like you were making my argument for me when drawing out the comparisons and the role luck plays in how our lives unfold. Yet, we draw radically different conclusions. It would be nice to know how to resolve disagreements at this level. I suspect James, Wittgenstein, and others are right in claiming that these disagreements boil down to psychology. There’s even some evidence for this from experimental philosophy. As should be expected, the findings are…complicated. :)


We agree about the normativity of terms like “parasite,” “infect,” etc. But in your initial post, it seemed to me you were trying to suggest that by using these terms, I, and others, am framing things in a “biased manner.” That toxo is counted as a parasite is a matter of biological taxonomy. That toxo has the ability to bring about epigenetic remodeling (via methylation) in the brains of mice is clear. That this can decrease the predatory aversion in mice (especially in relation to cats), is also clear (and fairly well understood). So, while “parasite” has a normative component, it also has a descriptive component as well. It’s not I am just throwing around the term “parasite,” just to score some kind of rhetorical points. I am calling toxo a parasite because that’s what the science suggests it is in relation to other organisms like mice. And to the extent that toxo has the capacity to influence human behavior (in ways that run counter to our interests), then it’s a parasite with respect to us as well. So, while I am happy to admit that to label something a parasite is to label it negatively with respect to something else—and hence the term is normative in that sense—it doesn’t mean I have to agree that by talking in these admittedly normative terms, I am framing the issue in a misleading or biased way.


I am not sure why it’s an “easy trap” to talk about control and manipulation when it comes to toxo. First, let’s imagine that humans develop mind-control devices that are inspired by the way that toxo influences mice behavior—namely, by “rewiring” our brains to influence our thoughts and behavior. With this device, scientists can make people attracted to rather than repelled by snakes, spiders, etc. Now imagine they use this device to invert a person’s aversions and attractions. Wouldn’t you claim that the scientists were controlling, manipulating, etc.? “Yes,” you’ll presumably say, “but that’s only because the scientists have minds, intentions, etc.” A few responses:

Why think only something with (conscious?) mental states can control or manipulate? This seems to generalize into an account that seems open to counter-examples to me.

Even if I grant for the sake of argument that mental states are necessary for control and manipulation, I would still like to hear how you would describe the relationship between toxo and mice (and perhaps us). Given that I am hypothesizing that the mind-control technology is built on toxo’s existing “technology,” it doesn’t seem like a trap, at least in this case and least to me, to talk about toxo as a manipulator of mice (and perhaps us).

That said, I wanted to briefly touch upon something else you said—namely, that I allegedly take the exogenous nature of the relationship between toxo and humans is somehow “conclusive for showing that agency is compromised in the toxo case.” But I wasn’t trying get from exogenous to agency-compromising in the earlier comment. I was trying to get from exogenous to alien. Something could be alien and non-compromising. So, while I do think a case can be made for treating toxo as both alien to us and agency-compromising (especially if one thinks agency comes in degrees), one need not run the two together. For my part, I think toxo could potentially play the role of an “alien force” when it comes to human decision-making and behavior (depending on what future science tells us about how it affects our thoughts and actions).

p.s. The point I was making earlier about contrast classes had to do with the importance of properly isolating the actual influence (if any) toxo has on human behavior. Here contrasts between those who are infected with toxo and those who are not are critically important. So, too, are contrasts between those who are infected and affected by toxo and those who are infected but not affected. I wasn’t trying to suggest, however, that these are the only contrasts that matter—only that other contracts might lead us further away from the target of our initial investigation.

John, you say: "These factors make some people much more inclined toward anger, fear, anxiety, and so forth, than others. And yet we hold them morally responsible... Why don't [the bugs, and their effects on us] just set up certain predispositions and inclinations, against which [these people] can still display the signature and distinctive human capacity for reasons-responsiveness?"

Well, this capacity isn't monolithic and can be significantly compromised by the same factors that make people more inclined toward anger, fear, etc. So the bugs and their effects on us, and genetics and early childhood experiences, could indeed "eliminate or etiolate our moral responsibility" if MR depends on reasons-responsiveness, it seems to me.

But as a skeptic about desert, I don't hold anyone MR, even the most reasons-responsive among us; I just hold them responsible. The moral bit is just our emotions demanding an additional justification for praise and blame beyond their natural consequentialist function.

"But as a skeptic about desert, I don't hold anyone MR, even the most reasons-responsive among us; I just hold them responsible. The moral bit is just our emotions demanding an additional justification for praise and blame beyond their natural consequentialist function."

Tom--with a slight protest against the use of "natural" this is very much what I am now about in terms of R, especially with regard to the function of emotions in claiming MR. Maybe we should thus distinguish between MR and R? My pragmatist tendencies are on show here.

Thanks for the clarifications, Thomas. I don't think the etiology of an action, or set of desires, or lack of aversions, is sufficient to get us to a conclusion about whether the subject was manipulated. It isn't even necessary.

Take a clear example. Iago manipulates Othello. He does this without using any alien methods -- he just uses Othello's own normal mental capacities. For what it's worth, I think this is the most ordinary use of 'manipulation', at least in the context of of manipulating persons.

The reason why I don't think toxo manipulate humans or mice is that they lack all of the pieces of Iago's manipulation of Othello, beyond having a causal effect on the subject's actions. They lack a plan, intentions, a proper conception of their actions in the context of their broader effects, etc. In short, I don't think they know what they're doing (at least at the level of human action). But I doubt they even know what they're doing at more microscopic levels because I doubt they think anything, which I think is a prerequisite for knowing thing. Not everyone agrees -- we're familiar with talking about a thermostat as an intentional system. I agree we can talk about thermostats in this way, but I don't think they are intentional systems. So maybe we disagree on this score.

Suppose Jones has a mutation that causes him to release certain pheremones that induce risky behavior in those around him (in a way similar to how toxo works). I think Jones doesn't manipulate those around him, nor do his pheremones, even if it makes those around him more reckless. And this is because manipulating another is, roughly, an intentional action. Or so I think. (But, I take it, we definitely disagree about this.)

Hi Thomas,

Thanks for your further reply.

I was, indeed, suggesting that the authors of the paper that touched off this whole discussion were labeling things in a biased manner. I didn't say that you were, but I did mean to suggest that those who wish to argue from (1) the claim that X exerts an influence on human behavior to (2) the claim that X exerts an *alien* influence on human behavior must take care not to describe X in a manner that begs the question. As you point out in reply to Matt, you are interested in making this argumentative move in relation to toxo. So I do think this is something that you should be careful about. At the same time, I am sure that you are more sensitive to the problem here than are the authors of the target paper.

The point I want to insist on is this. To label toxo a parasite at the initial step of describing it as an exogenous influence on our behavior (where the implication is that toxo is a parasite with respect to human beings [this may be more complicated than it appears to be, as evidenced by the Hamblin piece you link to in the Atlantic]) is to beg the question of whether it is an *alien* influence on our behavior. A parasite just is an alien influence. So the desired conclusion--namely, that toxo is an alien, exogenous influence on our behavior--simply falls out of the terms in which the argument is initially framed.

It would be better to describe toxo, initially, as an exogenous influence on our behavior (and perhaps, one could even introduce the idea that it is a parasite with respect to mice). And then one could argue that the influences toxo exerts on our behavior are such that it should be considered alien. My point is simply that this needs to be argued for and that the debate cannot fairly be presented in terms that assume it is true.

Now, it is hotly contested just what the sense of "self" in debates about our agency and free will should be taken to be. But I think it is uncontroversial that the relevant sense of "self" is not to be identified simply with the human body or nervous system. (This is one mistake that the authors of the original paper seem to make.) The agential self we all care about here is more nuanced than that. This is one reason why exogenous influences are not straightforwardly alien. And it also seems uncontroversial that the agential self operates under various influences, some of which are alien and some of which are not. So an influence on an agent's behavior is also not straightforwardly alien. This is why it takes argumentation to establish that a given influence on one's behavior, exogenous or endogenous, is alien. But it makes no sense to say that a parasite is not an alien influence. A parasite just is something that is understood in opposition to its host, as having competing interests. And so the argumentative work is done once one claims that a given parasite influences one's behavior.

Again, I did not mean to say that you were guilty of begging the question in this way. I was reacting to the abstract of the paper you posted (and some of the further reactions you posted as well). But I do think that this can be an easy trap to fall into and well worth guarding against.

The comments to this entry are closed.

Books about Agency

3QD Prize 2014: Marcus Arvan