A Defense of the Extended Mind Thesis
by KARINA VOLD
Abstract. In their article “The Extended Mind” (1998), Andy Clark and David Chalmers introduce a theory of extended cognition. In this paper I explain what extended cognition theories maintain by examining one such theory in particular- namely the Extended Mind thesis (EM), which Clark and Chalmers put forth. Following this, I consider two popular objections raised against EM- one based on concerns about what exactly constitutes a “part” of a cognitive system, and the other based on the intuition that the biological body is what marks the natural boundary between humans and their environments- and provide a defense of EM from each of these objections.
In the past, philosophers have thought of the biological brain and body of the agent as being the sole physical substrates that make up the mind. But, over the last decade a theory of the mind has emerged which suggests that a human’s mind, in particular one’s mental states and cognitive processes, may at times “extend” into the environment that immediately surrounds their body. According to the Extended Mind thesis (EM), parts located beyond the agent’s body can serve as the material vehicles of the agent’s mind and, in such cases, these relevant parts should be viewed as constitutive parts of the mind. In this sense, contrary to what has been traditionally thought, EM claims that the mind “extends” beyond the body.
This paper has three major parts: first, I will outline the central claims of EM and the arguments put forth by Andy Clark and David Chalmers in favor of it (section 2.1.) and follow this with an example of the theory (2.2.); second, I will consider two objections to the theory, one based on concerns about what exactly constitutes a “part” of a cognitive system and the other on the intuition that the biological body is what marks the natural boundary between humans and their environments; and third, I will provide I defense of EM against both objections, in each case arguing that the theory is able to withstand the objection raised against it.
2.1. The Extended Mind thesis
Andy Clark and David Chalmers first introduced EM in an article called “The Extended Mind”. In this article the authors make two central claims- one regarding cognitive processes and the other regarding mental states. Mental states include things such as propositional thoughts, experiences, beliefs, desires, feelings, and so forth. Cognitive processes, on the other hand, are processes that take place within an agent’s mind, or cognitive system, such as “retrieval of memories, linguistic processes, and skill acquisition.” The first claim EM makes is that an agent’s cognitive processes can be partially constituted by portions of the world that are not bound by their brain-and-body.
How I have phrased this claim is important since if, instead of “not bound by their brain-and-body,” I were to have said “non-biological” or “parts of their external environment, I would be begging the very question that EM is attempting to address. After all, the theory aims to redefine the very notion of what is internal or external to the agent. So, to say, for example, that an agent’s cognitive processes can be partially constituted by “parts of their external environment,” would be to assume that the relevant portions that lie beyond the agent’s body are external to the agent and a part of the agent’s environment, and thus, that they are not a part of the agent or the agent’s mind. Similarly, it would be wrong to use the phrase “non-biological” in this case since Clark and Chalmers remain open to the possibility that one agent’s mind may be partially constituted by the mind of another agent (further discussion of this in 4.3.). Thus, by claiming the mind can be realized by vehicles outside of the traditional “shell” of a human (their brain and/or body), EM is arguing that what is internal to a human agent is not (always) just what is internal to its body. After all, if parts of the world beyond an agent’s body really are the physical realizers of that agent’s cognitive process then these parts should be seen as internal to the agent.
The second claim of EM is that an agent’s non-occurrent mental states, such as beliefs and desires, can be partially constituted by portions of the world that are not bound by their brain-and-body. A non-occurrent belief is one that is not currently being entertained, whereas an occurrent belief is one that you are entertaining right now. For example, if I were to ask you whether you believed the sky to be blue in color, you would surely say yes. If so, then before I had asked you this question, your belief that the sky is blue would have been non-occurrent (unless you were already entertaining the thought). But, having been asked about your belief regarding the color of the sky, you likely brought that belief into consciousness and thus, what was once your non-occurent belief has now become your occurrent belief that the sky is blue.
The distinction between occurrent and non-occurrent beliefs is important as Clark and Chalmers are clear that the EM thesis only makes claims about non-occurrent mental state extension, and not occurrent mental state extension. This is likely because it is often held that our occurrent mental states have a phenomenally conscious feel to them, that is a feeling of “what it is like” to be in that occurrent mental state, but Clark and Chalmers both hold that consciousness cannot not be realized externally from the brain-and-body (although they are silent about why this is). So, in order to avoid committing themselves to the claim that consciousness extends, Clark and Chalmers limit their claims about extended mental states to non-occurrent ones. Furthermore, to avoid the same result with regards to their first claim, they must also limit the extension of cognitive processes to those that are not conscious.
Thus, for those who take the view that all cognitive processes and all mental states are conscious and occurrent, such as Galen Strawson, EM would be implausible since it only makes claims about mental states and processes that do not display consciousness. The view that all our mental states must be occurrent, however, does not seem plausible to me, and Clark and Chalmers would agree, because it denies the highly intuitive claim that one can have a belief, a desire, or knowledge that they are not currently entertaining. Still this is the position that Brie Gertler endorses in “Over-extending the Mind” (denying premise four of the argument for EM, see section 3.1.) and also what Clark and Chalmers’ believe to be “the most consistent way to deny” EM.
2.2. Otto and Inga: An illustration of EM
In The Extended Mind, Clark and Chalmers provide an example that attempts to draw a parallel between two people who want to go to the museum: one whose mind “extends” the other whose mind does not. The latter individual is Inga. After deciding that she would like to go to the museum, Inga quickly recalls that the museum in located on 53rd street, so she proceeds in that direction. In this instance it is clear that Inga has successfully relied on her working memory to access the information she needs to find her way to the museum. Meanwhile, a man named Otto suffers from Alzheimer’s disease and thus can no longer rely on his memory to recall all of his standing beliefs. Otto is forced to rely on a notebook that he stores all of his important information in such as phone numbers, directions, medical information, and so forth. After deciding to go to the museum, Otto quickly reaches for his notebook, reads that the museum is located on 53rd street and promptly heads in that direction. He relies on his notebook on a regular basis, takes it with him everywhere he goes, and writes in it often so that he will not forget important information.
Clark and Chalmers argue that “in relevant respects the (two) cases are entirely analogous: the notebook plays for Otto the same role that memory plays for Inga.”
In other words, if we consider the information in Otto’s notebook, then it seems that it plays the same explanatory role as the information in Inga’s memory. For example, in both cases we can explain why they headed toward 53rd street with reference to their desire to go to the museum and belief that it was located there. Inga’s belief was there before she accessed it (non-occurent) and likewise, Otto’s belief was in his notebook before he accessed it. To be sure, this is not the claim that Otto’s behavior is identical to Inga’s, but rather that “taken as a single integrated system, Otto-and-the-notebook exhibit enough of the central features and dynamics of a normal agent having… the dispositional belief (about the location of the museum) to warrant treating him as such.” Thus, a proponent of EM would grant that in many trivial ways Otto’s actions do differ from Inga’s, but still would contend that what is relevant is the role that the information about the location of the museum plays in each case, and in Otto and Inga’s cases, the information plays the same functional role.
3.1. A representation of an argument for EM
In this section I will flesh out each of the premises of an argument for EM, spending the most time on what I believe are the most important premises. In doing this, I will introduce and spell out both the ‘parity principle’ and Clark and Chalmers’ conditions of a cognitive system. Gertler has extracted the following argument from Clark and Chalmers’ Otto and Inga example:
- “What makes some information count as a standing belief is the role it plays.”
- “The information in the notebook functions just like [that is, it plays the same role as] the information constituting an ordinary non-occurrent belief.”
- The information in Otto’s notebook counts as standing beliefs. (1,2)
- Otto’s standing beliefs are part of his mind.
- The information in Otto’s notebook is part of Otto’s mind. (3, 4)
- Otto’s notebook belongs to the world external to Otto’s skin.
- The mind extends into the world. (5, 6)
3.2. Premise one and the Parity Principle
The first premise of this argument states that what matters in labeling some thing as a part of a cognitive system is not its location or physical identity, but rather the “role that it plays.” This premise follows from what Clark and Chalmers label the parity principle, which contends that if an object in the environment is playing the same role as an object that, were it located in the head, we would certainly count it as part of a cognitive system, then we should count the object in the environment as a part of the cognitive system also. So, if the information Inga’s mind (about the location of the museum) would surely count as a part of her belief (about where the museum is located) because of the role it plays in her memory, then, since the information in Otto’s notebook (about the location of the museum) plays the same role in his memory, it ought to count as a part of his standing belief as well.
If it weren’t the role that mattered but the physical identity, for instance, if the vehicles of cognition had to be biological (and within the body of the agent), then any non-biological resource, such as Otto’s notebook, would instantly be disqualified as cognitive. The parity principle is used to avoid this outcome. It maintains that regardless of how the information is physically realized, if the information in both cases plays the same functional role in driving the agent’s behavior, then it should be given the same cognitive status. After all, with no justifiable reason, it would be both ad hoc and question begging to assume a priori that only a certain type of matter- biological matter -can constitute standing beliefs.
So, it follows from the parity principle that whatever turns out to be the necessary functional role for a thing to play in order to count as a part of a cognitive system it will have to be described without reference to any particular physical properties and in a sufficiently abstract way as to allow that the relevant functional role could be played by something beyond the agents body. At the same time, the cognitive system cannot be described so abstractly as to include all sorts of things as constitutive parts of the system. Otherwise too many things would be included as parts and EM would threaten to “over-extend” the mind. For example, the information in Inga’s head should count as her standing belief not because it is biological or internal to her body, but rather because of the role it plays in her successful retrieval of memory, which then causes her to head towards the museum. This presupposes that there actually exist some way of describing the relevant “functional role” that is sufficiently abstract to be realized both internally and externally to the agent’s body.
3.3. Premise Two
The second premise of the argument for EM claims that the information about the location of the museum plays the same functional role in Otto’s case as in Inga’s case. To outline this functional roll Clark and Chalmers provide three (tentative) conditions to be met in order for any thing to be granted “recognition as part of the physical substrate of a cognitive system:”
- Constancy: The use of the resource must be a constant in the agent’s life.
- Accessibility: The resource must be directly and easily available.
- Reliability: The agent must trust and endorse the resource without hesitation; rarely doubting it’s veracity.
A fourth condition, which they are more tentative about labeling as necessary, requires that the information in the resource must be there as a consequent of having been consciously endorsed at some point in the past. It seems evident that these conditions are met by our ordinary, non-extended mental states and it is difficult to deny that Otto does not meet these conditions, since Clark and Chalmers developed the Otto and Inga example in order to support their argument for EM. Thus, it seems the best way to object to the second premise would be to claim that parts of cognitive systems, like mental states and cognitive processes, meet some further condition not on this list. The two objections that I consider in the second part of this paper take this approach in rejecting EM.
So, because the information in both cases meets these conditions and we would surely call the information in Inga’s mind her non-occurrent belief, then, given the parity principle, the third premise of the argument follows: we ought to call the information in Otto’s notebook his non-occurrent belief as well (P3). And, if the information in the environment serves as the material vehicle of a part of Otto’s beliefs (P3), and his beliefs constitute a part of his mind (P4), then it follows that the information in the environment is a constitutive part of Otto’s mind (P5). Furthermore, it is clear that something like a notebook is a part of the world external to an agent’s skin and body (P6). Thus, portions of the would external to Otto’s body (the information in his notebook) serve as the physical substrates of a part of his mind (namely, his belief about the location of the museum), and in this sense Otto’s mind “extends” into the world (P7).
4.1. Problems facing EM
My discussion in this paper will be limited to a consideration of two objections, both of which I will defend EM against. Neither of these objections are to the parity principle itself, but rather, they both deny the similarity between cases of extended cognition and regular cognition that is needed for the principle to apply. In other words, both of these objections object to either the necessity or the sufficiency of the functional role, as Clark and Chalmers’ have described it, in determining whether or not some resource is a part of a cognitive system. While I recognize that there are other important and compelling objections that one could make to EM, unfortunately I will not be able to consider all of them in this paper.
4.2. Sufficient conditions for “parts of cognitive systems”
The first objection I will consider would likely arise if one were uncomfortable with the results of EM, believing that it casts too broad a definition of what is mental. In this case they could object to the joint sufficiency of the conditions proposed for a cognitive system- constancy, accessibility and reliability (as discussed in 3.3.) This would require that there is some quality unique to all parts of cognitive systems that the three conditions fail to pick out. For example, if one could find a condition that is common only to our internal mental states, of the sort Inga has, and not present in Otto’s case, then the two cases would not be analogous and so, the principle would not apply. One quality that has often been thought of as being unique to biological cognitive systems is their capacity to produce intrinsic content. It is argued that this marks a distinction between the biological substrates that traditionally compose cognitive systems and potential non-biological realizers like Otto’s notebook.
This notion of intrinsic content rests on the view that mental states are intentional states, which is to say that they “have content; they are typically about things.” This is not unique to mental states, however, since certain non-biological things, such as books and road signs, also have content and so should equally be labeled intentional. So, according to this view, it therefore possible to draw a distinction between things with derived intentionality and things with intrinsic intentionality. Frederick Adams and Kenneth Aizawa define derived content as content that is assigned by “intentional agents who already have thoughts with meaning.” Non-derived content, or intrinsic content, on the other hand, does not require “the independent or prior experience of other content.” If something has content, then “it is either mental or the content is derived from something that is mental.” So, derived content is a product of intentional agents. It is argued that this marks an important variance between Otto’s case and Inga’s case- Otto’s notebook has derived content only and thus, it is argued that, its intentionality does not make it mental, whereas Inga’s memories are not derived, and therefore they do count as mental.
4.3. Responses to Intrinsic Content
In this section, I will argue that the objection based on the inclusion of intrinsic content as a necessary condition to be met by all parts of a cognitive system is actually no threat to EM. As Clark points out in “Intrinsic content, active memory and the extended mind”, the argument for intrinsic content is flawed for several reasons. First of all, the notion of “intrinsic content” is not uncontroversial and is surely not universally accepted. Secondly, Adams and Aizawa give no reason to believe that external, non-biological structures are (logically or contingently) incapable of having intrinsic content. We should not assume that this is the case and as Clark points out in the future we can imagine there might be technology that has advanced enough to change this reality. For instance, imagine that we identified some internal part as a vehicle of intrinsic content. Now imagine that part is replaced with a “functionally equivalent silicon part.” If the agent could still experience mental states with intrinsic content, then this should defeat concerns about intrinsic content as a criterion for mentality.
Furthermore, even if we grant the claim that intrinsic content is a necessary condition for being considered a part of the mind, EM would still be true in cases where one agent’s standing beliefs are realized by biological structures outside its body, for example, in the mind of another agent (who is capable of intrinsic content). After all, according to this objection it is only the fact that non-biological structures are incapable of intrinsic content that would prevents Otto’s notebook being recognized as a part of his cognitive system. Thus, there is no reason why the biological realizers of an agent’s cognitive system must be located within their own brain or body. For example, Clark and Chalmers suggest that “(i)n an unusually interdependent couple, it is entirely possible that one partner’s beliefs will play the role for the other as the notebook plays for Otto.” All that is necessary for the resource to count as a part of the cognitive system in this case is that it is capable of intrinsic content and meets the other three conditions laid out by Clark and Chalmers.
Finally, Clark’s third response is that even if intrinsic content is accepted as a valid notion and if non-biological structures were incapable of displaying it, this would not compromise EM. This is because in a system not every constitutive part of that system will necessarily have the qualities that the entire system, as a whole, has. According to Clark, all that is necessary for a given part to count as a vehicle of a mental state “is that it be appropriately linked… to representations whose content is (as Adams and Aizawa insist) intrinsic.” Adams and Aizawa make this very same point about the relationship between systems and their components themselves, only in a different discussion, when they attempt to clarify Clark and Chalmers’ positions in EM. Still, although they demonstrate that they are well aware of this relationship suggested by Clark, they do not seem to consider it when raising their objection to EM based on intrinsic content.
To spell out Clark’s claim I will use the example that Adams and Aizawa give of an air-conditioning system. In such a system only a portion of the components are responsible for actually cooling the air. The other parts play different roles- some “duct the air about the building, compress the refrigerant, directs the flow of refrigerant, monitors the room temperature, and so forth.” Thus, for an air-conditioning system to be labeled as such it has to cool air. Cooling air, then, is a necessary feature of an air-conditioning system, and yet, not every part of an air-conditioning system actually cools air. I would contend that, in the same way, original content may be a necessary condition for cognitive systems, but this does not entail that every material vehicle responsible for a given mental state must display original content. Furthermore, I do not see how even a defender of intrinsic content could deny this, since one would have to hold the same view with regard to brain-and-body bound mental states as well.
Surely no one would argue that every neuron and atom of grey matter displays intrinsic content just because the whole mental state that arises from a system composed of neurons and grey matter does. In fact Adams and Aizawa as much as concede this point themselves: “Having argued that, in general, there must be non-derived content in cognitive processes, it must be admitted that it is unclear to what extent every cognitive state of each cognitive process must involve non-derived content.” And, because we must remain neutral on where parts are located, as not to the beg question, we should apply the same standard to extended mental states. Thus, if EM is correct and mental states can extend into their environment, there should be requirement that external portions of the mental state be capable of intrinsic content in order to be recognized as constitutive parts.
5.1. Appeal to the dual boundaries of perception and action
Another objection to EM is based on the intuition that the natural boundaries that separate the mind from the external environment are the dual interfaces of perception and action and that these boundaries align with the boundaries of the body. In other words, the first claim is that our minds affect the world through our actions and likewise, our minds are affected by the world through our perceptions. If this is the case, then it follows that perception and action are the interfaces through which contact is made between our environment and us. The second claim of this objection is that these interfaces align with our body such that we perceive through bodily senses and we act through bodily motions. And, given these two claims, it follows that the mind must be body-bound.
For example, it is argued that because Otto must perceive the information in his notebook, this means that they will lie beyond his perception, and thus beyond the natural boundary that separates him from the world. Inga, however, does not need to engage in bodily perception to access her belief in the way that Otto does. In her case, her belief is realized by physical substrates located on the “inner” side of the perception and action boundary. So, because Otto must engage in this crucial act of bodily perception (and action) to access his beliefs, he would be taking in an extra step(s) that is not necessary in “ordinary” cases, such as Inga’s. What is more, because of this extra step that Otto takes, he would have a completely different phenomenological experience than Inga. If this is the case, then Otto and Inga’s cases are not entirely analogous and so the parity principle may not be applicable.
5.2. Response to the appeal to perception and action
There are a number of possible responses that a defender of EM could give to this objection. The first that Clark and Chalmers make is to point out that the objection begs the question. This is true, after all, it is the legitimacy of this very boundary- between an agent and her environment- that is in question and that EM is trying to define (as discussed in section 2.1.). So, a defender of EM might agree with the first claim of this objection- that perception and action are the dual interfaces that separate the mind from the world- but reject the second premise- that we perceive and act only through our bodies. After all, it is the second claim where the question begging is really taking place. By rejecting the second claim we could allow that the mind lies strictly on one “side” of our perceptions and actions, and remain uncommitted as to where perception and action occur.
Consider Walid, for instance, who is blind and relies heavily on his white walking stick to aid his mobility. Since the stick is his only reliable way to detect objects in his path, Walid uses it constantly and must keep it with him at all times. For Walid, this stick is more than just a helpful mobility tool- he uses it fluently and feels as though he were touching the pavement at the end of his stick. That is, he perceives the world at the end of the stick, not just his hand gripping the stick. I believe that in this case we can accept that Walid’s perception is not limited to his bodily senses, and thus, we can reject the second premise and conclude that (under the right conditions) cognitive processes can extend to include non-biological structures. As long as the process remains explanatorily similar in every relevant way to how a person with regular vision might function, then as the parity principle contends, we ought to label both instances as cognitive processes.
I believe this response to be compelling, since as Clark and Chalmers argue, it would be incorrect to allow which “side” of the body-boundary the process falls on to bias us. Thus, the example of Walid seems to help in the defense of EM, however, it does not work so effectively in defending the example of Otto and Inga. It doesn’t seem reasonable to say that Otto’s perception is not limited to his bodily senses because he perceives what is inside the notebook. In seems more likely that in Otto’s case, unlike Walid’s, the interfaces of perception and action do align with his body. Furthermore, the Walid example does not overcome the objection over Otto and Inga’s different phenomenological experiences either. After all, Walid would surely be having a different phenomenological experience of walking down a busy sidewalk than a person with 20/20 vision.
Clark and Chalmers’ second move is to downplay the difference between the phenomenological experiences of Otto and Inga. They do this by arguing that in both cases the agent will still have some phenomenological experience and, though the experiences may be different, this should not affect their equal status as non-occurrent beliefs. Chalmers expands on this idea by making an appeal to the notions of ‘introspection’ and ‘mental action’. According to him, if these notions can be taken as parallel to (‘real’) perception and (‘real’) action, respectively, in the traditional sense, then the parity principle will apply.
This response will work only if there are no substantial differences between the notions of introspection and perception, or between mental action and action. I take this parallel to be a little far-fetched myself and Chalmers recognizes that not everyone will be sold on this parallel, so he correctly predicts the possible replies that will follow. For example, one who opposes EM could insist that introspection does not seem to involve bodily sensory perception in the way that “real” perception does. And likewise, mental actions do not involve any physical action. Furthermore, one could insist that the differences in phenomenological experiences are too essential to be downplayed. Otto has a particular “perceptual and agentive experience” that Inga would not have, and so the cases are not analogous.
In this paper I have attempted to give a thorough exposition of the central thesis of EM- that the material vehicles that realize certain mental states and cognitive processes of a human’s cognitive system are at times located beyond the body of that agent. I have laid out the arguments put forth by Clark and Chalmers in defense of EM and explained the rationale behind them. In the second part of this paper I discussed two popular objections to the second premise of the argument for EM (as laid out in section 3.1.) and in each case I have argued that EM is able to overcome these criticisms. While I believe that I have successfully defended EM from these objections, I understand that, other serious objections still face the theory. But, unfortunately, I could not address them all on this occasion.
Adams, Frederick and Kenneth Aizawa. “The Bounds of Cognition.” Philosophical Psychology, 14 (2001) 43-64.
Adams, Frederick and Kenneth Aizawa. “Defending non-derived content.” Philosophical Psychology. (2004).
Aizawa, Kenneth. “Clark’s conditions on Extended Cognition Are Too Strong.” (2005).
Chalmers, David. “Foreword to Andy Clark’s Supersizing the Mind.” Supersizing the Mind: Embodiment, Action and Cognitive Extension. (Oxford, 2008) ix-xvi.
Chalmers, David. February 6, 2009. “Fodor on the extended mind.”
Clark, Andy. “Intrinsic Content, Active Memory, and the Extended Mind.” Analysis, 65 (2005): 1-11.
Clark, Andy. “Coupling, Constitution, and the Cognitive Kind: A Reply to Adams and Aizawa.”
Clark, Andy. Supersizing the Mind: Embodiment, Action and Cognitive Extension. (Oxford, 2008).
Clark, Andy and David Chalmers. “The Extended Mind.” Analysis, 58 (1998): 7-19.
Fodor, Jerry. “Where is My Mind?” London Review of Books. February 2009. Review of Supersizing the Mind: Embodiment, Action and Cognitive Extension by Andy Clark.
Gertler, Brie. “Overextending the Mind?” Arguing about the Mind. Brie Gertler and Lawrence Shapiro, eds. Routledge, 2007.
- Andy Clark and David Chalmers. “The Extended Mind.” Analysis, 58 (1998): 4.
- Ibid, 10. It should also be noted that it is preferable to use the term “agent” in this case instead of “human” in order to allow for the possibility that human minds may include other non-biological realizers (as when a blind man relies on a seeing-eye dog), or even that animal minds themselves may extend (if they can be shown to adequately engage in tool use, for instance).
- Clark and Chalmers, 6.
- Clark and Chalmers agree. Ibid, 4.
- Brie Gertler, “Overextending the Mind?” Arguing about the Mind. Brie Gertler and Lawrence Shapiro, eds. Routledge, 2007, 11.
- Clark and Chalmers, 9.
- Ibid, 6.
- Clark and Chalmers, 6.
- Andy Clark, “Intrinsic Content, Active Memory, and the Extended Mind.” Analysis, 65 (2005): 7.
- Gertler quotes the first two premises from Clark and Chalmers, The Extended Mind. Gertler, 2.
- Clark and Chalmers, 7. This is also what Gertler labels as premise one of the EM argument, 2.
- Gertler agrees, 7.
- Clark and Chalmers, 9.
- Andy Clark. Supersizing the Mind: Embodiment, Action and Cognitive Extension. (Oxford, 2008), 88.
- Gertler, 8.
- Jerry Fodor. “Where is My Mind?” London Review of Books. February 2009. Review of Supersizing the Mind: Embodiment, Action and Cognitive Extension by Andy Clark, 5.
- Frederick Adams and Kenneth Aizawa. “Defending non-derived content.” Philosophical Psychology. (2004), 1.
- Ibid 1.
- Fodor, 5.
- Clark, “Intrinsic Content, Active Memory, and the Extended Mind”, 4.
- Ibid, 4.
- Here we see an example of how functionalist reasoning is applied to support the EM argument.
- Clark and Chalmers, 10.
- Clark, “Intrinsic content, Active Memory, and the Extended Mind”, 4.
- Frederick Adams and Kenneth Aizawa. “The Bounds of Cognition.” Philosophical Psychology, 14 (2001), 50.
- Aizawa, 2.
- Adams and Aizawa, “The bounds of cognition,” 50.
- Clark and Chalmers, 9.
- The stick fulfills all of Clark and Chalmers’ three conditions- consistency, accessibility and reliability (see section 4.3.).
- Clark gives a similar example of a blind man, Supersizing the Mind, 31.
- Clark and Chalmers, 9.
- Chalmers, “Foreword to Andy Clark’s Supersizing the Mind,” xii.
- Ibid, xii.
- Ibid, xii.