INTEGRAL WORLD: EXPLORING THEORIES OF EVERYTHING
An independent forum for a critical discussion of the integral philosophy of Ken Wilber
Publication dates of essays (month/year) can be found under "Essays".
SEE MORE ESSAYS WRITTEN BY ANDY SMITH
AND HOLARCHIC NONSENSE
Andrew P. Smith
Jeff Meyerhoff recently posted at this site Six Criticisms of Wilber’s Integral Theory, which serves as a convenient summary of the main arguments in his book Bald Ambition. One of these criticisms, which he calls “Holarchic Controversy, not Holarchic Consistency”, quoted several passages from my own writings. As readers of my essays here will know, I am a strong critic of certain features of Wilber’s model, and I fully approved Meyerhoff’s inclusion of these passages. I want to take this opportunity to expand a little on Meyerhoff’s critique as it applies to holarchy.
Two Meanings, Many Quadrants
This is the kind of mess the four quadrant diagram leads to.
The passages Meyerhoff quoted make two critical points: 1) Wilber’s four quadrant model conflates the distinction between individual and social holons, on the one hand, with that between individual and social aspects of holons, on the other; and 2) Wilber inconsistently applies a criterion he uses to determine the hierarchical ranking of holons. I think the first point is fairly clear, and as I emphasize in the passages quoted by Meyerhoff, others have come to a similar conclusion. I just want to add here—also re-emphasizing a point also made in one of the quoted passages—that this is not by any stretch of the imagination a minor criticism. The four quadrant model can represent either individual and social aspects of holons, or individual and social holons—but not both.
If one chooses to represent individual and social aspects of holons—and the last passage of mine quoted by Meyerhoff refers to a Wilber statement implying that this is now his view—the four quadrant model can no longer be considered to represent, however generally or abstractly, all of existence. That famous figure that first appeared in the inside cover of Sex Ecology Spirituality, and which Ken continues to display (most recently in his soon-to-be-published Integral Spirituality) is out of the window. One can still make the distinction between individual and social holons, but one cannot portray their relationships diagrammatically. As I have pointed out in my recent An IMP Runs Amok, this makes it difficult to elucidate these relationships.
On the other hand, if Wilber wants to preserve the original meaning of this diagram, then he must dispense with the notion—repeated again and again and again in his writings—that the four quadrants represent four different aspects of holons. But this view of the four quadrants is central to much of his recent work, including Integral Methodological Pluralism (see my An IMP Runs Amok for a discussion of other problems with this methodology), and integral mathematics (which I have criticized in The Pros and Cons of Pronouns). This sense of the four quadrants is also essential to the elegant work Mark Edwards (Through AQAL Eyes, 1-7) has done in defining multiple perspectives of holons. So however one wants to define the four quadrants, the damage to Wilber’s model is quite severe.
It seems to me that the only move open to Wilber at this point is to embrace the conflation openly, and insist that his four quadrants can mean two quite different things. They can refer to individual and social holons, and also to individual and social aspects of holons. A single diagram can’t refer to both at the same time, but it can refer to one or the other, as specified. This is rather in keeping with his habit of defining terms in several different ways, and depending on the reader’s alertness to keep track of which definition he is employing at any one time.
To take this approach, though, is not simply confusing; it creates further problems. Suppose the four quadrant diagram is used to represent the aspects of a particular holon. This is basically the way Edwards now understands it, using a separate diagram to portray any particular holon. Wilber, however, says social holons don’t have the four fundamental aspects. So how many do they have, and how would these aspects be represented—on a two axis diagram?
On the other hand, suppose the four quadrant diagram is used to represent individual and social holons. The individual holons go in the upper two quadrants, while the social holons go in the lower two quadrants. However, the individual holons, according to Wilber, do have four aspects, and the social holons, one supposes, have a couple. So where do these aspects go in this kind of four quadrant diagram? Even if the four quadrants are now used to represent holons themselves, the aspects of the holons must have some relationship to these quadrants. But what is it? Specifically, what is the relationship of the social aspects of an individual holon to a social holon? And what is the relationship of the individual aspects of a social holon—if it even has such—to an individual holon?
This is the kind of mess the four quadrant diagram leads to. As I said earlier, this conflation between types of holons and aspects of holons ought to be fairly obvious. What continues to amaze me, however, is how Wilber, and his supporters, have lived for so long in denial of this contradiction. In an essay posted at this site that was very favorably reviewed by Wilber, Fred Kofman parroted these two contradictory views of the four quadrant model, practically in consecutive sentences, without showing the slightest awareness of the problem. I rebutted Kofman’s article within days of its posting, and a few months later discussed the individual vs. social contradiction at length. Neither Kofman nor Wilber ever responded. In my view, Kofman’s article is one of the worst ever posted at this site. It displays ignorance of the simplest, high school-level scientific facts, faulty logic, and a transparent political bias in attempting to argue for a particular view of the relationship of societies to individuals. Yet it was translated into other languages, and as I write now, more than five years after it was posted, it is one of a handful of essays thought worthy of being specially featured. Why? I can only hope because it is recognized as an object lesson in how not to defend theories.
Asymmetric Application of Asymmetry
What I show here is that the asymmetry criterion, as used by Wilber, does not work.
The second point made in the passages quoted by Meyerhoff—Wilber’s inconsistent use of a criterion for hierarchical ranking—is not so obvious, and I feel the brief passage Meyerhoff quotes to support it may leave readers unconvinced at best, confused at worst. It is the more critical point, because it goes to the root of why Wilber makes a distinction, in his four quadrant model, between individual and social holons in the first place. While I have been over this point in much greater detail in several of my previous articles (see, for example, All Four One and One For All, and God is not in the Quad), I think this is a good time and place to revisit the argument. What I propose to do here is go through this reasoning step-by-step, identifying the problems with it as I go along. This discussion may not always be simple and easy to follow, but I see no reason why it should be. Though I’m convinced Wilber’s logic is deeply flawed, he has obviously put a lot of time and effort into constructing his model, so it stands to reason it should take considerable further effort to untangle and bring to light the problems with it.
Any hierarchical model must begin with a criterion for distinguishing higher forms of existence from lower. The most widely-agreed criterion I am familiar with is emergence, the notion that higher forms of life have properties not found in lower. You will encounter this definition over and over if you read the works of complexity theorists and others who are interested in hierarchy. Wilber, however, eschews this approach, and instead uses what I refer to as an asymmetry criterion: “Destroy any type of holon and you will destroy all of the holons above it and none of the holons [more precisely, not all of the holons on any level] below it.” (SES, p. 61). This criterion is absolutely critical to his four quadrant model, because using it, he argues that while eliminating all individuals will eliminate all societies, the converse is also true. Thus he concludes that societies have a symmetric relationship with their individual members, that is to say, the two kinds of holons are on the same level of existence. This conclusion is at the root of his claim that individual and social holons require separate quadrants or axes.
The notion that societies are not hierarchically higher than their individual members contradicts the conclusion we would draw from emergence, for societies very clearly have emergent properties that none of their members have. So there must be something wrong with using one of these criteria. What I show here is that the asymmetry criterion, as used by Wilber, does not work. If applied consistently to all holons—something he doesn’t do—it leads to nonsense.
We begin by observing that use of this criterion depends critically on how we define holons. We have to know what we mean by “type of holon.” As applied to the individual/society relationship, this criterion only gives the result that it does for Wilber because he regards humans of different eras—e.g., magic, mythic and rational—as distinctly different types of holons, to be compared only with their corresponding societies. If one took the more logical approach—one I am sure would be taken by any biologist, sociologist or anthropologist—of considering all members of Homo sapiens as representing a single type of individual holon, then the asymmetry principle would indicate that societies are higher than humans, since the latter can exist in the absence of any societies.
Why then does Wilber regard humans of different eras as different kinds of holons? How does he rationalize this move? If one looks at his four quadrant model, it represents (in the upper right quadrant) the brains of humans of more recent eras as higher than those of earlier eras. In other words, Wilber believes that he can apply the asymmetry criterion to human brains of different eras, and demonstrate that they form a hierarchical progression. That being the case, he is justified in regarding humans of different eras as distinctly different kinds of individual holons. With each type of human isolated conceptually from the others, he can then apply that same asymmetry criterion to the corresponding societies, showing that their relationship with their individual members is not asymmetric.
For example, he can argue that we moderns have a symmetric relationship with our society, because if that society were eliminated, so would we be eliminated. It is not that the elimination of all modern societies would eliminate all human beings, but that it would eliminate all modern human beings, who are now defined as a class or kind of holon different from humans who might have lived in some other kind of society. The same logic can be applied to humans of other eras. Since humans in any particular kind of society are defined as a different kind of holon from humans in another kind of society, by that very definition eliminating that kind of society will eliminate that kind of individual.
In the preceding paragraph and a half, I have used the term “humans”, but keep in mind that Wilber’s argument is based not on human organisms, but on human brains. This is where he begins to apply his asymmetry criterion. This is a very peculiar way to construct a model that purports to represent all of existence. The fundamental unit of biology is, of course, the organism, not the brain, but nowhere in Wilber’s four quadrant model do we see organisms actually represented. Not to make the organism a fundamental holon (or in Wilber’s terms, a fundamental exterior aspect of a holon) in a hierarchical model must be considered a serious deficiency.
Why does Wilber base his upper right exteriors on brains rather than on organisms? His logic is based to a considerable degree on evidence that the human brain includes the brain of lower organisms, in the sense that lower portions or regions of our brain are somewhat similar to the entire brains of these other organisms. All of us have, so to speak, the brains of lower animals, plus something more that they lack. Thus we have the autonomic brain, the reptilian brain, the limbic brain and the cerebral cortex. This is the triune brain concept, with our brain really consisting of three brains (Wilber adds to this the autonomic brain, which is justified on the same grounds that the other brains are). In contrast, the human organism does not include lower organisms, in the sense that the human brain includes the brains of lower organisms. The human brain may, in an approximate sense, include the brain of a bird, a reptile, and an invertebrate, but the human organism does not include any of these organisms.
However, Wilber then extends the triune brain concept to assert that the brains of humans of more recent eras transcend and include those of earlier eras in the same way that any human brain transcends and includes the brain of a lower animal. He has no anatomical evidence for this claim whatsoever. There are no studies demonstrating that the brain of modern humans is anatomically distinct from the brain of humans of earlier eras, let alone distinct in precisely the same way that any human brain is distinct from those of other animals. That is to say, there is no evidence that a modern human brain contains all the structures of an older human brain, plus some additional structures unique to it.
Wilber surely must be aware of this, so why does he nevertheless maintain that such a relationship exists? I believe his reasoning is as follows. While it has not been demonstrated that a modern human brain includes, let alone transcends, the brain of an earlier member of our species, he can argue with a somewhat greater degree of plausibility that much of our behavior transcends and includes that of earlier humans. Thus in Wilber’s model, the rational human is capable of various cognitive processes that earlier humans were not. This being the case, it must also be the case that these unique processes have correlates in unique brain structures.
I think most scientists would agree with Wilber that if there truly are class differences—that is, differences shared by all members of a group, as opposed to individual differences—in behavior between humans of different eras, these differences must be reflected in class differences in brain structure. But it is a huge leap of faith on Wilber’s part to presume that these hypothetical structural differences are such that a modern human brain can be said to transcend and include an earlier human brain, or that the asymmetry principle indicates that the former is higher than the latter. In the case of the triune brain, which evolved layer by layer, we can see that the lower structures were essential for the development of the higher. But lacking any direct evidence that the modern human brain is composed in the same manner of earlier human brains, we can’t say anything about the necessity of the latter.
So at the very least, we must conclude that arranging Wilber’s exterior quadrants hierarchically depends critically on behavioral, rather than anatomical, evidence. We have no evidence that, at the human level, structures are hierarchically arranged. Nevertheless, some of Wilber’s supporters might consider this enough. Mark Edwards, in particular, has argued strongly that exteriors include not simply physical structures like brains, but the behavior manifested by these structures. To the extent that we understand exteriors in this manner, it appears that the right hand quadrants can be viewed as hierarchical.
However, even if we view them in this way, there is still another problem with the logic here: it isn’t applied consistently, that is to say, to other, lower levels. Does a neuron transcend and include, say, a heart cell, a lung cell or a liver cell? It doesn’t appear to. But if we consider a never cell simply in terms of its behavior—just as we have seen Wilber must define human exteriors, if they are to be hierarchically arranged—we could say that it does. A nerve cell has the ability to communicate with other nerve cells not simply by direct, physical contact—as do other kinds of cells—but by indirect, synaptic interactions. The latter is a very different and arguably higher form of communication, just as modern human communication, which involves indirect (books, internet, TV, etc.) as well direct interaction, is higher than earlier, direct human modes of communication. This difference, moreover, is clearly asymmetric. Synaptic interaction requires cells that have the earlier, direct type of interaction, but the latter does not require synaptic interaction.
Still further, in the case of the nerve cell, there is no problem demonstrating that its distinct behavior is correlated with a distinct structure. A nerve cell differs from other, non-excitable cells in the molecular properties of its surface membrane, and in other ways as well. This unique structure is also related asymmetrically to the structure of other cells. Other types of cells can and do exist in the absence of the neuron’s unique features. But these unique features presuppose the more basic properties that other cells have.
So if we apply Wilber’s asymmetry criterion to neurons in the same way that he applies it to humans and other organisms, we must conclude that neurons are higher than other kinds of cells. That being the case, we can then go on to apply the same criterion to determine the nature of the relationship between individual neurons and the brain. If we eliminate all neurons, we clearly eliminate all brains. But the converse is also true. If we eliminate all brains, we eliminate all neurons. Neurons do not exist independently of brains. Thus we must conclude, following Wilber’s logic, that the brain is no higher than any of its member neurons.
I could extend this argument to other kinds of cells. We could carefully examine the properties of different kinds of cells in the body, and conclude that we can hierarchically arrange them according to their behavior. Again, this would allow us to conclude that no tissue or organ is higher than its component cells. I could even make an analogous argument on the molecular level . The properties of atoms depend to a large degree on the molecules they are found in. Atoms in small molecules like amino acids have properties that the nominally same atoms do not have when they exist free.. The same atoms have still more complex properties when found in more complex molecules, like proteins. So we could argue that atoms are also hierarchically arranged, and that in each case the atom is at the same level of existence as the molecule that contains it. The result, as I have discussed elsewhere (Intersubjectivity, Interobjectivity and the Collapse of the Four-Quadrant Model), is a reductio ad absurdum. The hierarchy is effectively collapsed to a few levels.
Let me summarize the discussion. We have seen that Wilber defines hierarchical relationships not by considering the emergence of new properties, as is conventionally done, but rather by an asymmetry criterion. Eliminating all holons on a lower level will eliminate all holons on higher levels, but not the reverse. One problem with this approach is that his criterion leads to different conclusions from that of emergence. Using the latter criterion, we can clearly identify several hierarchical stages within cells and also within organisms, which are completely ignored by Wilber’s approach. It isn’t simply that his model treats these relationships differently; it’s that his model doesn’t even recognize them.
Wilber uses the asymmetry criterion to argue that a) humans of different eras are on different levels; and b) humans of any era are on the same level as their society. His argument for a) is based on evidence that there is an asymmetric relationship between the human brain and that of other organisms. But there is no evidence that the brains of humans of different eras are related in this manner. No one has demonstrated that the brain of a modern human being is anatomically distinct, let alone in an asymmetric manner, from that of the brain of an earlier human.
Therefore, Wilber’s asymmetry claim, as it applies to humans, must be based on behavioral differences. That is, he must rely on evidence that our behavior could not exist in the absence of earlier forms of behavior, while the converse is not the case. However, a consistent application of this approach reveals that there are also asymmetric relationships among many lower holons, as depicted in his model. For example, neurons are higher than other types of cells, based on the observation that their behavior depends on properties found in these other cell types, while the behavior of the latter does not depend on the unique properties of neurons. Having made such a distinction between neurons and other cells, we can then consider the relationship between neurons and brains. Using the asymmetry principle, we conclude that neurons and brains are on the same level of existence, since neither can exist without the other. We can perform a similar analysis on other cell types and on lower holons. Doing this results in a collapse of the hierarchy, in which most holons are concluded to be on the same level. This is a reductio ad absurdum.
Wilber’s flawed use of the asymmetry criterion, then, is what leads him to view societies and individuals as occupying the same level of existence. There are numerous other lines of evidence that are inconsistent with this conclusion. To name just one, complexity theorists have shown that human societies, neuronal networks in the brain, and metabolic networks within cells all have a fundamentally similar kind of organization. This suggests that networks of neurons and metabolic networks should be considered the most appropriate lower level analogues of human societies. They are considered so in my model, and this is a direct consequence of understanding that social holons are hierarchically higher than their individual members, as both kinds of networks have emergent properties. In contrast, the lower level analogues of societies that Wilber’s asymmetry criterion identifies, such as stars and planets, are not characterized by such interactions.
1. Both of these systems in fact also conflate the individual/social distinction. Nominally, the multiple perspectives they define belong to either individual or social holons, but on closer examination some of the social holon perspectives are also social aspects of individuals. For example, the third-person perspective is defined diagrammatically by Wilber as the view of a society, but he also uses it to refer to the view of a particular individual member of a society. Perhaps this is the root of his conflation. He assumes that any individual third person view is identical with the view of an entire society.
2. Wilber, as far as I can tell, does not flatly deny that emergence is not a characteristic that can distinguish higher holons from lower ones. He also does not deny that societies have properties that none of their individual members have. His point, at least the argument I have heard some of his supporters make, is that individuals likewise may have properties that societies don’t have. The common (in fact only) example is a visionary, who develops new concepts that go beyond those currently recognized by society. But even visionaries are members of societies, and depend on this membership both to formulate and to disseminate their ideas. So any individual advance is also a social one. In fact, there are many individual properties that are not shared by all members of society, so the fact that some properties are quite rare, even unique, does not mean that they are not properties of the society as well as the individual.
Moreover, there are numerous other examples of holonic relationships where the asymmetry criterion does not agree with the emergence criterion, and where there is no possibility of claiming a symmetrical distribution of emergent properties. Consider molecules. In Wilber’s four quadrant model, all molecules exist on a single level of existence. Yet some kinds of molecules, by the criterion of emergence, are clearly higher than others. Proteins have properties not found in their individual amino acids, and proteins are components of still more complex holons with their own emergent properties. Based on emergence, we could (and in my model I do) distinguish several hierarchical levels or stages between the simplest molecules and cells.
Applying the asymmetry criterion to these stages, though, is problematic. Cells clearly cannot exist without molecules, but most molecules found within living cells cannot exist independently of these cells, either. They depend on the cell both to synthesize them and to protect them from degradation. So based on a strict application of the asymmetry principle, one would have to conclude that almost all the molecules within cells are on the same level of existence as the cells themselves. How in the world would this be portrayed in a four quadrant diagram?
Another example is provided by tissues and organs in the body. Again we can distinguish, by the criterion of emergence, several distinct stages or levels of multicellular holons, ranging from very simple tissues to complex organ systems. And again, if we use the asymmetry criterion, we have to conclude that all these holons are on the same level as the organism, because just as the organism can’t exist without these tissues and organs, so can the latter not exist in the absence of the organism.
There presumably was a time during evolution when amino acids existed without proteins, proteins existed in the absence of more complex molecules, and the most complex molecules existed in the absence of cells. Likewise, the evolution of organisms must have been preceded by tissue-like organizations of cells. So depending on how strictly the asymmetry principle is applied, one could argue that it does distinguish these different stages of molecules and of multicellular holons. However, since Wilber’s model does not recognize these stages, we would then have to conclude that he applies the asymmetry criterion selectively, using it to distinguish some higher/lower relationships, but not others.
3. It might be argued that humans are by nature social creatures, and can’t exist without some social arrangements. If one defines society as any kind of interaction between organisms, than this is true. But obviously what we moderns commonly view as societies are not essential to individual humans. The fact that even today we observe an enormous spectrum in social development attests to the ability of humans to live—if conditions make it possible or necessary—with very little interaction with others. At the very least, we can conclude that none of the civilizations/societies that have existed on earth for the past ten thousand years was necessary for the survival of individual humans—that their demise would not have necessarily meant the end of the human species—and therefore did not exist in a symmetric relationship with them.
4. Notice, however, that the relationship of these different brains to each other is quite different from the relationships of lower holons that, in his model, are considered hierarchical. At these lower levels, such relationships all involve one holon containing a number of lower holons, usually a very large number. For example, a molecule contains several, often a great many, atoms; any cell contains millions of molecules; an organism (or brain) contains thousands or millions of cells.
This one-includes-many relationship has some very clear-cut consequences. For example, a cell is totally independent of any of its individual molecules; eliminate one of the latter and it doesn’t affect the cell in the slightest. The same for an organism and its cells. This, to me, is one of the cardinal features of genuine transcendence—complete independence from any single lower holon.
When we come to the higher brains, though, we find that each level of the triune brain contains just one of the lower holons; thus a reptilian brain contains one autonomic brain, a limbic brain one reptilian brain, and so on. This is very different from the one-includes-many relationship of higher to lower levels in other parts of the hierarchy. The triune brain is not independent of each of its lower brains; it could not survive, let alone function normally, in their absence. This is one reason why, in my model, a distinction is made between transformation and transcendence; both are higher/lower relationships, but of different natures.
So here we see another problem with Wilber’s definition of levels. The relationship of these brains is very clearly not the same as the relationship of higher to lower levels in other parts of the hierarchy. This ought to raise a warning flag: that the relationship of these brains needs to be represented somewhat differently from relationships like a cell to its molecules, or an organism to its cells. In other words, the asymmetry criterion lacks the power to distinguish some higher/lower relationships that are clearly different from others.
5.It is, of course, a trivial matter to demonstrate that the relationship between any human brain and that of any other organism is not simply much greater, but of an entirely different character, from that between a modern human brain and that of an earlier individual. Numerous studies of other intelligent animals, such as non-human primates, have shown that there are limits to how far they can develop behaviorally. Despite living in the most enriching environments, and exposure to human trainers, non-human primates can’t progress past a very low stage of human cognitive development. In stark contrast, any human being raised in such an environment can develop more or less to the same extent as any other. This demonstrates that the brain of any normal human has the same basic structural equipment as the brain of any other human. Depending on the society the individual is a member of, his brain may take on one of various states, or sets of states, but all these states are potentially available to any individual.
6. Edwards (2003,2004) The Depth of the Exteriors, Parts 1-3. But Wilber himself, it seems to me, places much of behavior in the left quadrants. For example, he regards our use of language—which is obviously a prime example of behavior that distinguishes modern humans from earlier ones—as an interior aspect. Also, as Edwards himself laments, Wilber rather consistently describes exteriors in terms of structures rather than behavior. He seems to intend his exteriors to be physical forms. And indeed, he has little choice, because if structures like the brain are not exteriors in his model, what else could they be?
7. Of course, neurons can exist in lower, different kinds of brains, the limbic, reptilian, autonomic. But I could also show that the neurons that exist in these brains are different from each other behaviorally (as well as, to some extent, physiologically). So we can construct a hierarchy of different kinds of neurons, just as Wilber constructs a hierarchy of different kinds of humans. Each type of brain is composed of a different kind of neuron, and in each case, he brain is no higher than the individual neuron. Indeed, in my model, the brain/neuron relationship is a lower level analogue of the society/human relationship.
8. That one cell type is ranked lower than another does not mean that it can’t have properties that the higher cell type does not have. In fact, neurons lack properties that specific types of other cells have. But exactly the same point can be made about humans. The fact that moderns may exhibit behavior that people of earlier eras did not does not mean the latter did not have their own unique forms of behavior. This is another reason why in my model humans of different eras are not related by transcend-and-include. My purpose here, however, is not to discuss how different cell types may be hierarchically related. It’s to show what happens when Wilber’s logic, as applied to human domains, is consistently applied to lower levels of existence.
9. The hierarchy is also collapsed because none of these cells and tissues can exist independently of the organism, and none of these atoms and molecules can exist independently of cells. See footnote 2. So effectively all molecules within cells are at the same level as the cells, and all cells within organisms are at the same level as the organism. Since atoms are also at the same level of their molecules, all forms of existence, from atoms to organisms, are at the same level.
10. Barabasi (2001) Linked (NY: Perseus)