Summary of and Comments on Chapter 8 "Metaphor and Truth"
of George Lakoff's and Mark Johnson's Philosophy in the Flesh
by Roger E. Bissell
submitted to MIND internet discussion group on 10/21/99


Lakoff and Johnson maintain that, despite appearances to the contrary, the controversies over the nature of reason and metaphor are not obscure pedantry. Instead, they say, "the philosophical stakes are...high": if their claim is correct that "the traditional view of metaphor is empirically false," then a number of philosophical theories and basic commonsense views of the world are at risk. In contrast to the traditional theory which (they allege) must hold (and usually is seen as holding) that metaphor is irrelevant to such issues as the nature of reality, truth, language, knowledge, and morality, L&K claim that views (both commonsense and philosophical) on such issues are necessarily tied to one's view of metaphor. If one's position on the latter is false, then any of one's views derived from that position "must also be false."

Comment: this seems obviously fallacious. Sound logic based on one or more false premises may generate a false conclusion (e.g., All cows are fish, all fish are plants, therefore all cows are plants), but it may instead generate a true conclusion (e.g., All cows are fish, all fish are mammals, therefore all cows are mammals). This is in contrast to the fact that true premises connected by sound logic must generate a true conclusion (e.g., All cows are mammals, all mammals are animals, therefore all cows are animals).

(An aside: we could look at the entire syllogism as being one large compound proposition: "all cows are mammals because all cows are fish and all fish are mammals." Since this is false, the conclusion, if taken as including "therefore" is false also. However, I am wanting to look at the truth-status of the proposition that is stated as the conclusion, first in abstraction from the fact that it has been generated by a syllogism (which means that the "therefore" and the specific premises are mentally set aside) -- and then with regard to whether there is a necessary connection between its truth-status and its having been generated by this particular syllogism. When we do this, we see (from independently validated knowledge that cows indeed are mammals) that the proposition generated by the premises is true -- and we see that there is no necessary connection between its being true and its being generated by those premises. In other words, its truth as the conclusion of those premises is accidental, i.e., not necessary.)

In other words, L&K could legitimately have said: if one's position on metaphor is true, then any of one's views derived from that position (in conjunction with other true premises and by means of sound logic) must also be true. Not exactly the direction that L&K want the logic to go!

This much is a pretty straightforward application of traditional Aristotelian logic. Since truth is a major concern in this chapter (and others), and since the issue keeps coming up in Objectivist discussion groups and usually is not fully clarified, however, I offer this brief additional digression.

As a tool of thought, the syllogism is an instrument or device for necessarily producing a true conclusion from true premises whose terms are related in the appropriate way. However, any syllogism that produces a true conclusion from one or more false premises whose terms are related in the same appropriate way only does so accidentally. As a proposition, the conclusion "all cows are animals" is true regardless of whether the premises you derive it from are true or false, but it is only necessarily true -- i.e., as a conclusion -- if the premises you derive it from are true. So, there is a crucial distinction here that is often overlooked: a true proposition that is a sound conclusion from one or more false premises, while a necessary conclusion from those premises and a true proposition, is not necessarily true as a conclusion.

Similarly -- and apropos of my comments about L&K -- a false proposition that is a sound conclusion from one or more false premises, while a necessary conclusion from those premises, and a false proposition, is not necessarily false as a conclusion. The upshot of this is that L&K can elucidate implications between the traditional theory of metaphor and various metaphysical and epistemological views until the cows and chickens come home, but they cannot claim to know that the latter are false based on the falsity of the former!

The most L&K ought to be saying here, if they were appropriately careful in their thought processes, would be something like: "...the views of reality, truth, language, knowledge, and morality that are tied to the traditional theory of metaphor may also be false." [Now, I should exercise appropriate caution here myself. It may be that L&K see these views of the world as being "tied" to the theory metaphor by some means other than logical implication (e.g., as corollary ideas, as conjoined with the theory of metaphor, etc.). If so, it is not clear to me what that "tie" is; to the contrary, what leaps out at me over and over during my own reasonably attentive reading of this chapter are the words "implies" and "entails."] Indeed, I think some/many of those views of the world are false, but that is something that will have to be taken up issue by issue, which is what we can do as we progress through the remaining parts of the book.

The flip side of L&K's argument, of course, is that IF the traditional theory of metaphor is correct, and IF that theory does in fact imply various of these views of reality &c, then those views must be correct, also! So, there's a lot more work to do in assessing all these logical relations and truth values than L&K seem to realize....

Central Points:

The guts of chapter 8 is L&J's examination of the five core beliefs of the traditional theory of metaphor (TTM) and why each of those beliefs is false -- along with an analysis of the widely accepted "objectivist" over-extension of the commonsense theory (CST) of truth and language, which validly applies to basic-level concepts and "embodied understanding," but not all truth and language. L&K are on solider logical grounds here.

They say that since the objectivist overextension of CST (let's call it CST+) implies the first four core beliefs of TTM, then if TTM is false, so also is CST+. (Since the relation between CST+ and TTM5, the "similarity" thesis, is somewhat different, I'll address L&K's comments on it after dealing with the first four.) So, as we examine the failings of tenets 1-4 of TTM, bear in mind that these failings are held to also invalidate CST+ (and any other application of CST beyond our basic-level concepts).

First, let's examine CST+ and its relationship to TTM's first four core beliefs:

1. CST+: ideas and thought have to be literal if they are to fit the world; therefore, ideas cannot be metaphorical; therefore, TTM1: metaphors must not be conceptual; any metaphoric use of a word must not fit the world and express truth, but instead refer to something else.

2. CST+: the function of ordinary language is to express ideas that fit the world as it really is; language that does not do this is not ordinary language; (from 1. above) metaphors do not fit the world; therefore, TTM2: metaphors are not part of conventional, ordinary language and normal thought.

3. CST+: the proper senses of ordinary words must be their world-fitting senses; (from 1. above) metaphors do not fit the world; therefore, TTM3: any metaphoric use of a world must not be proper but instead "deviant", meaning something other than what one is saying.

4. CST+: any apparently metaphoric world-fitting expression must not really be metaphoric, even if it once was; therefore, TTM4: metaphors can only be world-fitting if they are actually literal expressions, i.e., no longer real, "live," but instead now only "dead" metaphors.

Because of the above claims, widely accepted in our culture, metaphor is banned from and not taken seriously by "truth-seeking enterprises." So say L&J. And I think they're correct. In any case, they set out to show that tenets 1-4 of TTM are mistaken and thus that CST+, the non-embodied, "objectivist" over-extension of the common sense theory, is false. And if non-embodied CST is false, then the embodied version, limited to our basic-level concepts must be true!

So, what about TTM 1-4?

Contra TTM1: the Love is a Journey example shows how this tenet is false. Each of the linguistic expressions "are not simply distinct, different, and unrelated metaphorical expressions," but instead "are all instances of a single conceptual metaphor." What is primary and what unites all the related linguistic expressions is the "conceptual cross-domain mapping," i.e., the metaphorical thought.

Comment: L&K are absolutely right-on about this.

Contra TTM2: the expressions in the Love is a Journey example are part of our ordinary language, because the conceptual metaphor mapping Love into Journey "is part of our ordinary everyday way of conceptualizing love and reasoning about it."

Comment: You go, L&J!

Contra TTM3: the Love is a Journey example "is one of our normal ways of conceptualizing love," not a deviant one.

Comment: Right again.

Contra TTM4: the Love is a Journey example is live, because "it keeps producing more examples of new metaphorical expressions" and is "used regularly without awareness or noticeable effort." Thus, it is both conventional and a "live" metaphor. This is also true of the example of the Ideas are Objects metaphor and its corollary Understanding is Grasping, both metaphors still being very much alive. (An interesting sidebar: although both "grasp" and "comprehend" used to mean both "hold tightly" and "understand," only "grasp" still means both, while "comprehend" now means only "understand.")

(An aside: admittedly, L&J didn't analyze this example as well as they could have. However, I contend that everyone has some preferred sensory modality in which their linguistic metaphors tend to cluster. Elgin discusses this in her "gentle art of verbal self-defense" books, in which she explains how a lot of failure to communicate is due to people operating from a different sense-based system of metaphors. Your next communication breakdown really might be from something as simple as your preferring sight and your partner preferring tough or hearing. So, anyone disagreeing with L&K's analysis experiences "grasp" as not a metaphor but ordinary usage should realize that they are speaking for themselves and others whose sensory preference is not touch, but sight or hearing. Do you see what I'm saying? Does it sound correct to you? The error is to try to universalize from one's own preference on this. I think we all need to be aware of this factor operating sub-verbally or implicitly in our communication, so that we can adjust when appropriate for better understanding -- and so that we can have a better attitude toward people who are different from us. And check out Elgin's books, if you haven't already. If she is all wet (touch metaphor) in what she's saying, surely some bright (sight metaphor) persons can set me straight (touch metaphor).)

(Another aside: It might be thought that the knowledge as grasping metaphor has roots far back in the history of philosophy. Would you believe: Plato and Aristotle? I suspected as much, when I skipped ahead in L&K to chapter 18 on Aristotle and read their claim that he used the metaphors The Mind Is A Container, Understanding Is Grasping, and Ideas Are Physical Objects with a structure all their own. I got curious about this and consulted my old stand-by, Mortimer Adler's Syntopicon (encyclopedia of 102 key ideas from the Great Books of the Western World, 1952). In Adler's entry on "Knowledge," he wrote: "...a second point about the nature of knowledge which seems to be undisputed. If knowledge relates a knower to a known, then what is somehow possessed when a person claims to have knowledge, is the object known. It does not seem possible for anyone to say that he knows something without meaning that he has that thing in mind. 'Some sort of signal,' James writes, 'must be given by the thing to the mind's brain, or the knowing will not occur -- we find as a matter of fact that the mere existence of a thing outside the brain is not a sufficient cause for our knowing it: it must strike the brain in some way, as well as be there, to be known." What is not in any way present to or represented in the mind is not known in any of the various senses of the word 'know.' What the mind cannot reach to and somehow grasp cannot be known. The words which are common synonyms for knowing --"apprehending" and "comprehending"-- convey this sense that knowledge somehow takes hold of and surrounds its object." (p. 881) Plato in Meno wrote of the need to "fasten" or "bind" ideas in order to keep them from "walking" or "running" away. When I first read this my freshman year of college (1966), I thought it was rather droll. It takes on a new significance now, of course. Aristotle more clearly exemplifies the point made by Adler. In Metaphysics Book XII, chapter 7, he said that thought is active when it possesses (rather than just receiving) the object of thought. In On the Soul Book III chapter 8, he said: "The soul is analogous to the mind, for as the hand is a tool of tools, so the mind is the form of forms and sense the form of sensible things." And in Nichomachean Ethics Book VI chapter 1, he said: "We said before that there are two parts of the soul -- that which grasps a rule or rational principle, and the irrational; let us now draw a similar distinction within the part which grasps a rational principle. And let it be assumed that there are two parts which grasp a rational principle -- one by which we contemplate the kind of things whose originative causes are invariable, and one by which we contemplate variable things...Let one of these parts be called the scientific and the other the calculative...Therefore the calculative is one part of the faculty which grasps a rational principle." So, Aristotle explicitly used the grasping metaphor. Plato was close to doing so himself, it appears. It might be interesting to wade through all the Syntopicon references on the relation between knower and known to see which other major thinkers used this metaphor. But it seems likely to me that it is an undisputed feature of the Western view of knowledge, as Adler claims, and continues as a wholly conventional (though non-literal) metaphor to this day, not least in Rand's writings.)

L&J contrast live metaphors with linguistic expressions that "came into the language long ago as a product of a live conceptual metaphor," but whose "conceptual mapping has long since ceased to exist" -- such expressions being the only kind that qualify as dead metaphors, in L&J's opinion. L&K cite the example of "pedigree" which means "foot of a grouse," the image of which was mapped onto family-tree diagrams. Neither the linguistic nor the conceptual facets of this mapping still exist; the facts that all ties between the word and its original conceptual source have been broken and that the word no longer generates new, novel expressions that manifest the mapping together are why L&J consider this particular metaphor to be "dead," as against the Love is a Journey and Understanding is Grasping metaphors. Of course, this is yet another way in which L&J are bucking the Establishment (or their version of it), but they won't get any argument from me on this!

Comment: I've written about this previously and pointed out that L&J's criteria for what counts as "alive" or "dead" in re metaphors make a lot more sense in factual, functional terms than do such standards as whether an expression is now considered conventional, normal usage. (Here, parenthetically, is the gist of that discussion: claims that "grasp" is no longer metaphoric, but literal, are premised on the traditional theory of metaphor (see PitF, pp. 122-126), which L&J object to, of course. Tenet 2 says that "Metaphorical language is not part of ordinary conventional language. Instead, it is novel and typically arises in poetry, rhetorical attempts at persuasion, and scientific discovery." Tenet 3 says that "Metaphorical language is deviant. In metaphor, words are not used in their proper senses." Tenet 4 says that "Conventional metaphorical expressions in ordinary everyday language are "dead metaphors," that is, expressions that once were metaphorical, but have become frozen into literal expressions." In each of these respects, the traditional theory of metaphor would hold that "grasp" = "achieve knowledge of", i.e., that the latter is now a literal (not metaphorical) meaning of the former. But that is a perfect illustration of what L&J are trying to supplant. In particular, the whole idea of referring to a metaphor that is still generating meaningful associations as being "dead" seems clearly contradictory. Thus L&J give a better reference for what it would be for a metaphor to truly be "dead," an example of which is "pedigree" (see p. 124). (This, too, is a metaphor -- calling a metaphor "living" or "dead" -- and L&J are working more in the true spirit of those concepts.) So, Knowing is Grasping is a metaphor, even though (1) it is not novel, but part of ordinary conventional language, (2) it is not deviant, but a normal way of expressing the act of knowing, and (3) it is not literal, but (however apt) a conventional expression. If this is a paradigm shift from a well-entrenched view of metaphor, it is one that I think is well worth making. It is conceptually clearer and does less violence to our (largely metaphoric) understanding of what it means for something to be "living" or "dead." Webster's no doubt affirms that "grasp" = "comprehend" = "understand" is a widely accepted, conventional, normally understood, non-deviant meaning of the term. But it is a mark of the traditional theory that this is claimed to also be literal and non-metaphorical, i.e., a "dead metaphor." As L&J persuasively argue, its metaphorical function is not cognitively dead. (This reminds me, in an obscure and macabre fashion, of the recent attempts by neuroscientists to replace the traditional, loose notion of "viability" based on a fetus's lungs and heart being sufficiently developed that it is able to survive premature birth, to the more precise criterion of "fetal brain viability", where it has begun to have discriminated awareness and distinct waking and sleeping brain states characteristic of adult human beings.))

Another example: It has been pointed out that the term "flourishing" is a stand-in for living well, the metaphor I take to be that Living Well is Flowering. There may indeed be people who have grasped the idea of living well being flourishing, without also being aware of the horticultural roots (!) of the term "flourish." Unless the term was learned as an arbitrary sound assigned the meaning "living well," I don't know how the metaphoric association could not happen. For a living organism to live well is for it to vigorously expand and grow into a fuller/the fullest manifestation of its potential as a living organism. For a plant to flourish or "flower" is for it to vigorously expand and grow into a fuller/the fullest manifestation of its potential as a plant. "Plant" and "flourishing" or "flowering" are more concrete, closer to the basic-level concepts than are "living organism" and "living well." So, the latter maps into the former, the conceptual metaphor being Living Well is Flourishing. Now, suppose a person has never heard about flourishing in the context of plants, but nevertheless learns to associate the word "flourishing" with living well. If he/she has any sense at all of the aptness of this word for referring to a good life, it must be through another more concrete use of the term, namely, a flourishing -- i.e., expansive, self-extending -- kind of gesture. Again: to Live Well is to Flourish (i.e., to act expansively, self-extendingly). Anyone who does not get his/her understanding of the term "flourish" from some such metaphoric association is just learning a word-sound. But this does not invalidate the fact that the metaphor is still very much alive -- any more than it does, for instance, in regard to the fact that some people are opaque to the Love is a Journey metaphor. Metaphors are not for everyone! And not all metaphors are of interest or value to a given person -- unless he/she is a metaphor-ologist!

Finally, consider TTM5: there is an established and systematic similarity between the metaphoric and the normal, literal use of a given word. If a metaphoric word or phrase seems, in a particular context, to mean something other than its literal meaning, it must mean something about the subject matter of the sentence that is similar in some way to its literal meaning.

Contra TTM5, L&J give four arguments: (1) there is no preexisting similarity between the source and target of a cross-domain mapping (e.g., between Journey and Love); the similarities are not discovered but "created" by the process of one's choosing to relate the source and target. (2) there is no literal similarity expressed by the metaphor Knowing is Seeing, even though both involve knowing; (3) whereas similarity is a symmetric concept, metaphors, based on a target-source distinction and a one-way mapping, are rarely reversible; a particular target idea can take on the meaning of a particular source idea in a number of languages, but not vice versa; we naturally draw on images from the source domain in talking about the target domain, but not vice versa; and we use forms of inference original to source domains in thinking about target domains, but not vice versa; (4) since more than one metaphor can apply to a given idea, the potential exists for contradictions between them, which rules out preexisting similarity, but allows for conflicting mappings which "need not be simultaneously activated."

Comments: (1) the first argument seems to be obviously false. Aren't similarities, based as they are on real features of the things being compared, discovered, not created? For what possible reason would one choose to metaphorically associate two ideas, unless one, on some level of awareness, noted more than one similarity between them that prompted one to try to knit certain of their characteristics together into parallel systems of relationships? Even if the linkage is originated by "intuition," rather than by direct mental inspection, is one not acknowledging that which already exists, i.e., the parallel relationships between the two conceptual domains? I think the resolution to this conundrum is to carefully distinguish between a parallel and a similarity. The fact that pre-exists any metaphoric linkage is the two sets of features that parallel one another and that, once they are realized to be (or, perhaps, suspected of being) sets of features constellated in parallel fashion, are then seen to be similar in regard to their occupying corresponding positions in the constellations of features. In other words, I am hypothesizing (along with L&J?) that what is first grasped in conceptual metaphors is the overall pattern, the Gestalt, and the detailed, component similarities grasped in a later stage of development of the metaphor. So, while not literally "created" by the mapping, the mapping (Gestalt-grasping) does make possible the discovery of the similarities. (This argument is strictly off the top of my head, and I welcome better explanations, pro or con.)

(2) I think L&J are wrong here, too, as is apparent from the examples they give for Knowing is Seeing. "Murky" can generally mean "difficult to grasp clear boundaries and relationships among objects," whether mental or physical -- in which case, there is another metaphor operating: Seeing is Grasping, as is also the case for "I see what you mean." (I think a strong case can be made that the primary mode of cognition is the tactile sense, and that vision builds on touch.)

(3) Everything that L&J say about the asymmetry of metaphors is correct, but I think they are simply wrong about similarity being symmetric. By their line of reasoning, if one blue object is larger than another blue object, then they are only similar if the second blue object is also larger than the first! The error seems to be that L&J are confusing the mapping with the features between which the mapping occurs. The mapping is possible because the features can be seen (and are, once they are Gestalted) as similar.

(4) I think L&J are arguing a strawman concept of "similarity," that requires there to only be similarities or dissimilarities, but not both, between two things in reality. Clearly, both can exist. All that is required to meet the test of noncontradiction is that two things not be both similar and dissimilar in the same respect! Not 25 pages earlier, L&J explained how "contradictions" exist on different "levels of embodiment" (the neural, the phenomenological, and the cognitive unconscious), because they are distinct levels (i.e., respects) which are all equally real, none privileged metaphysically.

The same is true, without the need for an appeal to similarity theory, of the fact that marriage, for instance, can in one respect involve a business-partner relationship (putative equals) and in another respect involve a parent/child sort of relationship (unequals). In other words, both similarities exist; they do not wipe each other out, metaphysically!

(An aside: A particular marriage could be like a business partnership at one time and/or in one respect, while it could also be like a parent-child-relationship at another time and/or in another respect. The paradigm of contradiction is usually set up with one particular thing, but the same argument could be made about the universal marriage, i.e., about one of its instances contradicting one of its others by being different, when in fact it is just another case of difference at a different time/respect, so there is no actual contradiction.)

So, after the dust settles, L&J don't appear to have a leg to stand on in their objection to the similarity theory of metaphor. They say that if it is true, then two parts of CST+ must hold: (1) "all concepts must be literal and must name objectively existing things and objectively existing categories in the world; (2) "similarity must be defined by shared properties that really exist objectively in the world." Since I see both of these conditions as holding, without question --viz., since I see general kinds of things and characteristics being every bit as real and existent as individual things and characteristics,

I can't see any escape for L&J. If they are to have a solid footing for their view of metaphor, they will have to make a rapproachement with the fact that reality is objective! In particular, they will have to come to terms with the fact that not only concrete-level similarities exist, but also similarities of all kinds and levels of specificity and generality. The world is a very rich place, and we will never be able to fully "mine" all of the relationships that are there, waiting for us to discover them -- be they similarities, causal primacies, internal and external, explanatory primacies, etc. In their shying away from this fact, L&K are more like the philosophical tradition than they might care to acknowledge.

However, this is only part of CST+, and it does not appear to have fatal consequences for L&K's challenge to TTM. Indeed, if my comments above are correct, their critique of TTM is decisive, leaving standing only that portion of TTM that their model of metaphor must reconcile with (viz., the similarity theory), if it is to be compatible with the Objectivist (as against the "objectivist") theory of concepts.


Thus, L&J have set the stage for parts 2 and 3 of their book by considering several simple examples that, they say, reveal many radical implications for philosophy of their theory that "ordinary, everyday reason can be metaphorical," including:

1. primary metaphors which develop unavoidably and unconsciously out of our everyday experience act as links between our perceptual experience and our introspective experience and conceptual judgments; this is why our abstract concepts have the internal logic, imagery, and qualitative feel of perceptual experience.

Comment: although this point seems crucial to L&K's case, I'm not sure it is necessary in order to explain why our abstract concepts have perceptual "feel" aspects; after all, if our abstractions are properly tied to concrete reality by being correctly formed, that itself would establish the basis for the sensory "feel" of higher level concepts. On the other hand, if in some way L&K's model is an equivalent or "corollary" explanation of what is going on, and is truly complementary with the Objectivist model of concept-formation, then it actually gives more richness and perspective, rather than being in conflict with Objectivism. How could we settle this issue, as to the conflict or complementarity of the two models? This seems like a wonderful avenue for exploration and discussion...

2. abstract concepts have two parts, the literal skeleton and the metaphorical flesh, and the former is not rich enough to function as a concept without the latter; thus, many (all?) abstract concepts are defined largely in terms of multiple metaphors, which are often inconsistent with one another.

Comment: a red flag goes up here with the word "defined." Strictly speaking, definitions of concepts, whether more concrete or more abstract, cannot be and are not even allowed to try to be internally inconsistent. I do understand what L&K are getting at here, and the best source to refer to here for those who do not, is the series of books by Suzette Hadley Elgin with titles that are variations on "gentle art of verbal self-defense." She discusses how people often clash in discussions of ideas when they have differing definitions of one or more concepts, and the best way to tease out the hidden conceptual conflicts is to brainstorm a list of the associated ideas each person has with the concept in question. Often the two parallel lists will include not only identical/similar items, but also one or more items that are obviously antithetical to one another. Multiply this sort of largely tacit disagreement across a culture, and you have a recipe for the kind of conceptual clashes that have marked the course of ancient and modern philosophy. That is what L&K propose to examine, and I look forward with fascination to what they have to say. The idea that these philosophical conflicts have a metaphorical root -- or even are capable of being described in terms of competing metaphors -- seems to be a very fruitful one, one that could give a great deal of illumination to our grasp of the history of philosophy. It looks to me like a first-cousin of the conflict-resolution method Elgin uses to juxtapose the differing sets of component ideas and associations people have for a given word/concept. It's kind of a blend or synthesis of cognitive psychology and neuro-linguistic programming, if I understand all of this correctly. Certainly, it's the kind of area that we MIND-minded folks would find enlightening to explore.

3. the basic function of metaphor is to connect a "source domain" with a "target domain" with inference patterns, which is why reasoning is metaphorical.

Comment: this seems like a technical analogy to how concepts "hook" into one another by being nested in an ascending/descending series of genus/species relationships. Again, I think it is an area to explore for complementarity to, and not necessarily conflict with, the Objectivist epistemology.

4. abstract scientific theorizing is impossible without metaphorical thought.

Comment: I think this is true. It's hard for me to imagine what such theorizing would be like, and I can't imagine enjoying the task of trying to follow and understand it. This seems to me to be one of the strongest points that L&K make in support of their overall thesis.

5. metaphorical concepts require that the classical correpondence theory of truth be replaced with the view of truth as embodied.

Comment: L&K are saying that there is no such thing as "the truth", to which our propositions correspond, even if we don't realize it. Instead, truth is utterly dependent upon context and understanding. If there were a "single unified metaphysics," independent of mind and body, then the distinct truths that we obtain from different levels of "embodiment" of our concepts (e.g., the phenomenological level vs. the neural level) would clash, and there would be no way to decide which of them (e.g., phenomenologically, color inheres in things in the world, while neurally, color inheres in the relation between objects and our sense organs and nervous system) is true, i.e., which corresponds to reality. We need the additional fact of "levels of embodiment" to help resolve this dilemma. This seems correct, for it appears to be a special application of the law of contradiction: e.g., while a color cannot be both inherent in things in the world and not inherent in the world at the same time and in the same respect, that color can be both inherent in things in the world and not inherent in the world in different respects. Whether we think of these respects as different "levels of embodiment" or as what is grasped by different "forms of awareness", it seems clear that we must grant (i.e., acknowledge) the equal reality of each of them.

6. since formal logic is disembodied, literal, non-imagistic, and non-metaphorical, it is unable to say anything substantive about the nature of human concepts and reason.

Comment: I take this to mean that logic cannot prescribe content of any theory about human knowledge. In agreeing with this, I also want to emphasize that logic, being an abstraction from how all valid knowledge works, does prescribe the form or method by which any such theory is formed. So, to the extent that a theory of methodology has "content," it would be excluded from L&K's generalization. It is an instance of the fallacy of the stolen concept to claim that the law of contradiction, for instance, cannot exercise veto power over any cognitive theory that proposes actual contradictions can exist and play a functional role in our acquisition of knowledge.

7. reason and concepts are not independent of the body, but instead are shaped by our organs and the ways we function in the world.

Comment: since a process of thought and a process of forming concepts requires some sort of physical substrate in order for it to proceed (and by that, I mean that since all knowledge is derived physical contact between a knowing being and some aspect of the real, physical world), this hardly seems controversial, but I don't mean to underestimate the extent to which some sort of Platonic view of reason and knowledge still influences philosophy, especially via religion. To the extent that essences are based on real relations between more and less causally primary aspects of things, essences are metaphysical; this is so even though which (relatively primary) causal primary you select as a thing's essence is inescapably tied to the context of knowledge within which you are forming the current definition. So, while the material from which we construct a definition (argument, concept, etc.) exists independently of our body and manner of operating, the form in which we frame that definition, &c., including the specific material used for the definition, is captured by our bodies and our manner of functioning. This seems like an important and not altogether obvious distinction to keep in mind.

8. much basic metaphysics grows out of metaphor.

Comment: "grows out of" may be inaccurate, to the extent that metaphysical ideas can be framed with a process that does not necessitate a metaphoric link to our concrete (or less abstract) experience; however, it's clear to me that, at the very least, much basic metaphysics relies heavily on metaphor in order to make the ideas intelligible and convey them with much of a sense of reality, once they are framed, even if that sense is not always accurate.

L&K explore these claims further in the following chapters, where they show how "our most fundamental concepts -- time, events, causation, the mind, the self, and morality -- are multiply metaphorical." These concepts are so metaphorical, they claim, that even if we could somehow eliminate all use of or reference to metaphors, these concepts would be stripped of nearly all usefulness for basic thought. Philosophy simply cannot exist without a massive infrastructure (my word) of metaphors -- nor can science or abstract human thought in general. Yet, this isn't a tragedy, they say, but instead is a great intellectual gift, "the very means by which we are able to make sense of our experience."

Despite the strong exception I have taken to L&K's playing fast and loose with logical implication in arguing against the traditional theory of metaphor, I am in basic agreement with their overall thesis. It makes complete sense to me, and none of the attempted counter-arguments and refutations I have seen to date convince me otherwise.

Postscript comments: I agree that there is a need, and potential, for resolving the
apparent gap between Ayn Rand's and Lakoff/Johnson's in how our thoughts, ideas, and concepts are formed. One thing that may be helpful would be to survey Rand's writings, especially her epistemology monograph, and see where metaphor pops up the most, or the most strikingly. A hunch: I'll bet that Rand uses metaphor a lot more in discussing concepts of consciousness than existential concepts. For instance, she employs the/a traditional definition of knowledge as a "grasp" of reality. This clearly is metaphoric and is derived/borrowed from the most concrete level of action there is: the tactile grasping of physical objects. If you consider that she says such concepts (concepts of consciousness) are derived from our awareness of the world, it's not surprising that she (and the traditional view) sees (another metaphor!) knowledging as being a kind of "grasping." How many more of these metaphors are there in her philosophy? And how far does this idea of concepts of consciousness being derivative of concepts of existence go in explaining the use (even necessity) of metaphors in philosophy?

It is my prediction/guess that metaphysics and epistemology and ethics -- even Ayn Rand's! -- borrows heavily from (or builds extensively upon) basic-level concepts of objects, actions, characteristics, and relations; and that this tendency will only compound when we are expressing actions and relations from the realm of consciousness. One very important thing that verifying this would prove -- or at least illustrate -- is the utter necessity of using metaphors in doing metaphysics and epistemology, even for someone whom some regard as being very literal and precise (in the sense of saying what she means, so that you don't have to guess or wade through obscure hidden meanings to get what she's saying). L&J claim that if the use of metaphor, both conscious and (especially) unconscious, were denied to philosophy, very little of interest could be said. It seems clear to me that they are right. Nonetheless, there are Objectivists who are very uncomfortable with this idea.

I would personally be better able to spot all the metaphors in Rand's (or anyone's) writing once I finish L&K's book, especially sections 2 and 3. But even at a cursory glance, they leap out at me (heh-heh) from chapter 4 of ITOE. Paragraph 2: "On the lower levels of awareness", "integrate sensations into percepts". Paragraph 3: "Extrospection is a process of cognition directed outward...introspection is a process of cognition directed inward...", "the various actions of a consciousness can be experienced, grasped, defined or communicated." That's just the first page of the chapter, and I'll bet I've missed some metaphors at that....

Out of curiosity, I also looked at chapter 1 of ITOE and found quite a few metaphors:

p. 5, para. 1 -- process that consists of two essentials (whole-part) para. 2 -- consciousness develops in three stages...the base of all man's knowledge is the perceptual stage para. 4 -- in the form of percepts that man grasps the evidence of his senses...the perceptual level para. 5 -- the building-block of man's knowledge is the concept of an "existent" cannot grasp it explicitly until...

p. 6, para. 1 (partial) -- man grasps it...grasps the constituents of the concept "existence", the data which are later to be integrated by that concept (two metaphors: grasp and the part-whole metaphor) para. 6 -- the key, the entrance to the conceptual level...other living species are unable to follow p. 7, para. 1 (partial) -- the concept "unit" is a bridge between metaphysics and epistemology

That's enough. I hope that is sufficient to convince anyone who might have been doubtful about how much Rand (supposedly a very literal thinker and writer) relied upon metaphor in order to illustrate and clearly convey her ideas. And should we really be surprised, considering how well Rand was able to use imagery and symbolic levels of meaning in her novel writing? Metaphoric thinking came very naturally to her.