On Life and Flourishing
A Thesis
Presented to
The Division of Philosophy, Psychology, Religion & Linguistics
Reed College
In Partial Fulfillment
of the Requirements for the Degree
Bachelor of Arts
The Author of this Blog
May 2007
Acknowledgements
Thanks to Paul Hovda for his advising, and to Gavin Lawrence and Margaret Scharle for assisting in the research of this thesis.
Preface
David Lewis once wrote in his book On the Plurality of Worlds “Why believe in a plurality of worlds?-Because the hypothesis is serviceable, and that is a reason to think that it is true.” In the same book, Lewis writes, “Nowhere in this book will you find an argument that you must accept the position I favor because there is no alternative. I believe that philosophers who offer such arguments are almost never successful, and that philosophers who demand them are misguided. I give some reasons that favor my position over some of its close alternatives. But I do not think that these reasons are conclusive…” I regret that my reaction to Lewis’ theory at the time was not more sympathetic.
In working with the account of Aristotelian categoricals presented in Michael Thompson’s The Representation of Life, I hounded myself with questions as to what reasons we have for believing that an unrecognized form of thought exists. My work would perhaps have been more focused if I recognized the wisdom in Lewis’s account, and focused more on discovering ways in which Thompson’s hypothesis might serve to explain other true judgments we take ourselves to make. For I now think Thompson’s arguments may provide insight into a wide range of philosophical topics.
Why believe in Aristotelian categoricals? Because the hypothesis is serviceable. We say some things occur by accident, and other not. Aristotelian categoricals serve to explain thoughts about accident in ways different from modal concepts. Likewise, we sometimes are tempted to make strong claims about health, sickness, mutation, and what is normal, all of which can be accounted for by Aristotelian categoricals. Further, we use adjectives in describing organisms, a use not explained by our predicate and quantified logics. Given certain Aristotelian categoricals, a new explanation of such attribution is open for exploration. New perspectives on thinking itself are given by such a hypothesis.
Still, the theory is new and perhaps has unacceptable hidden implications. Perhaps the accounts I give in this essay might be better solved in different terms. Perhaps a theory with less commitments will come along and solve the same problems. These are worries that I cannot address yet, but I hope to deal with as they come along. All I ask of my readers is to hold onto their doubts about why they should believe in the kinds of thought proposed by Thompson until they have been given a clear picture what such a hypothesis enables.
Table of Contents
Chapter One Aristotelian Categoricals 1
Aristotelian categoricals’ grammatical and logical form 5
How Do Life-Form Propositions Interact With Other Propositions? 11
Accident Schema 12
Interference Principle 14
Modality 16
Coincidence and Chance 18
Interpreting the “c” Operator 22
Further Relation? 23
The Evaluative Schema 23
The Converse Schema 25
Attribution 26
Folk Biology 27
Chapter Two Epistemology of Life-Form Propositions 31
The Need For An Epistemology 31
The Challenges Of Giving An Epistemology 32
Argument Via Schema and Converse Schema 33
Argument Via The Interference Principle 36
Foot’s Method 41
Chapter Three Defect and Teleology 47
Limiting the evaluative schema’s scope 47
Teleology As a relation of Life-Form Propositions 48
Is Teleology Within or Between Life-Form Propositions? 50
Ends and Final Ends 51
The Final End? 52
The Modified Evaluative Schema 53
Why Do Teleological Relations Matter? 55
The Structure of a Defect Claim 57
Chapter Four The Normative Power of Natural Defect 59
The Rational Cuckoo Objection 59
Does Reason Radically Influence Norms? 64
The Choosing to Value Objection 66
Closing Remarks 73
Appendix A: Proposed Inferential Rules 75
Bibliography 77
Abstract
In this paper, I explain what Aristotelian categoricals are and critically analyze their use. In chapter one, I characterize Michael Thompson’s Aristotelian categoricals and then suggest that they be understood as a kind of proposition, which I call life-form propositions. I argue that the primary inference from a mismatch of life-form proposition and a regular proposition is accidence understood in a certain way, and that accidence understood in this way entails an explanation of a certain kind. I will then go on to characterize other forms of inference using life-form propositions.
In chapter two, I consider whether there are any special epistemological problems regarding life-form propositions and attempt to characterize such problems. I then argue that the commitment to explanation given by an accident claim allows the problems I find to be resolved satisfactorily. Philipa Foot’s argument regarding the epistemology of life-form propositions is then analyzed. While I find her critique of Thompson’s evaluative schema sound, a critique which serves as motivation for her account’s formation, her method of analyzing Aristotelian categoricals’ truth is in some ways unexplained and is too restrictive.
In chapter three, I consider the conditions needed to meet a general criticism argued by Foot to be required for inference to defect from life-form proposition mismatch. I then argue that Foot’s criticism can be met if the scope of the evaluative schema is limited to teleologicaly related life-form propositions. I then attempt to give an account of such teleological relations.
Finally, in chapter four, I consider three broad positions one might take on the evaluative strength of defect claims. I then present objections to these positions, and explain how a proponent of natural defect might respond.
Dedicated to those who supported me through college.
Chapter One
Aristotelian Categoricals
Michael Thompson, in his essay “The Representation of Life”, argues that thoughts deemed ‘natural historical judgments’ have a unique form distinct from usual thoughts about individuals or groups. Thompson dubs sentences that express these thoughts ‘Aristotelian categoricals’. Natural historical judgments are of the ‘wider context’ needed to experience some material as a life-form, or some activity as a life-activity. As one needs to be thought of as playing baseball for one’s actions to be ‘pitches’ or ‘battings’, so Thompson thinks thoughts about life-form are needed to experience an object as a living thing, and to think of its activities as life-activities such as eating, growing and reproducing. To understand an activity as an eating is to think of it as playing a part in a life, so the association of an activity under some life-form is necessary in order to think of such activities as ‘vital activities’. For instance, Thomson writes “In thinking of something as an acorn, we tie it up with oaks, none of which need be present here, and so of course we ‘look beyond’ the individual lump of stuff.” Here, Thompson is arguing that some form of thought needs to exist in order to experience the individual lump of stuff as an acorn, a form of life.
Thompson’s view is striking in that it treats association of an object under a life-form as necessary for the experience of the particular object as a living being. His theory thus opposes the view that certain signs such as complexity, having genetic material, being self-moved, metabolizing, and other such characteristics present criteria for thinking something living. The apposed view might hold that finding general features held in common by all living beings allows one to reach a concept of life. Since Thompson thinks a vital description of a life-form is needed before an individual thing can be even treated as a living-being, he appears committed to denying the previous view. This commitment is expressed in the claim that “Vital descriptions of individual organisms is itself the primitive expression of a conception of things in terms of ‘life form’ and ‘species’...If this is right, then, of course, we are wrong to think of the concepts of the various life-forms as reached through abstraction from features of their particular bearers.” This line of thought also applies to objects such as individual cats and the species cat. Thompson appears to hold that the species cat should not be thought of as an object of thought reached by looking at individual cats and abstracting to some general feature they hold in common, because a concept of the life-form cat is already employed in the very process of thinking of some individual thing as a cat.
One interesting case Thompson presents in discussing his views on vital descriptions is that of the sudden creation of a human-like swampman. In the hypothetical situation proposed, lightning strikes some mud in such a way as to make a perfect physical duplicate of Thompson. Thompson argues that only if swampman is thought to have the same life-form as he do we believe swampman to be alive. For when the context given by human life-form is dropped, problems occur. One problem is determining if the material on the swampman’s “side” is really an arm. Such a problem does not occur for the author Thompson, as it is said of Thompson, in light of his form, that the things on his side are arms. The same cannot be said of the swampman. While the swampman’s ‘arms’ are physically exactly like Thompson’s, we cannot say that the mud-thing has arms, for there is no ‘wider context’ or vital description for the mud-thing that tells us that the objects connected to it are arms and not mutilated wings, or deformed legs. This also holds true for activities such as eating; it is only as a life-form that Thompson eats, rather than just a process of chemical reactions occurring.
A slightly similar case raises doubts about the importance of the difference between Thompson and swampman. Suppose, rather than lightning striking some mud, a quantum accident occurred in a woman such that a swampman “zygote” “grew”, “developed”, and “gave birth” to a thing physically duplicate to the reader, that then went on to have the same experiences as he did. If such a case is possible, the reader might be a swampman. Such an accident seems possible, and if it occurred, Thompson seems committed to denying that the reader is alive for the same reasons he denies the swampman a life-form. This example is interesting for it questions whether having arms, legs, and eating in the quoted, figurative sense, is really less important to us than the non-quoted, actual sense. On the face of it, this example shows that having arms in the quoted sense is not important to us, for if it turned out we have actually been a swampman our whole life, then the importance having arms has to us is really just the importance of having “arms”. As such, Thompson’s concept of vital activities do not seem to pick out something more important to us than what is had without such form.
One way out of this situation available to Thompson is to treat thinking itself as a vital activity. Thompson in fact seems to make something like this line of argument in his recent essay “Apprehending the Human Form”. If thinking is a vital activity, Thompson is in the position to give a kind of cogito argument. From the perspective of the person who doubts whether he is a swampman or not, the thought “I am thinking” is enough to determine that the person is not a swampman. For, from the premises “I think” and “Swampman does not think”, the person can conclude that he is not a swampman. Thus, to his skeptic, Thompson can deny that he could possibly be a swampman, and thus deny the possibility of the scenario.
The justification for this argument relies in part on the assumption that thinking is a vital activity available only to life-forms, or even perhaps life-forms of a certain kind. Further, the argument assumes that swampman is not of some life-form, and hence will not think, only act as if he does think. In other words, the argument assumes that we can only know about swampman qualitatively, through observing his behavior and actions, and not through swampman’s inner experiences. This assumption is in conflict with our previous hypothetical situation, for in it we assumed that our lives would be the same if we turned out to be swampman. So, we assumed swampman could have some inner experiences, as we have thoughts, feelings, desires, etc, and we thought we could have been a swampman. This may be a flaw with the hypothetical situation, for what we know about some swampman seems to come from observing his behavior, which seems to preclude us from knowing if he has the experience of thinking, since such knowledge seems like only self-knowledge. So Thompson could deny that the scenario we are thinking of is one in which we are actually a swampman, as a swampman could not have our inner experiences.
Aristotelian categoricals, the sentences expressing such vital descriptions as discussed above, can thus be thought of as expressing different aspects of some life form. A given Aristotelian categorical will be about some species and will express something about that species. For example, we say such things as “tabbies eat mice”, “tabbies mate in the spring”, and “tabbies have orange and white fur” in discussing the lives of tabby cats. These sentences can serve as Aristotelian categoricals about tabby cats, and express something particular about what it is to live as a tabby cat. Collections of Aristotelian categoricals about tabby cats form a ‘natural history’ of the tabby cat in that they describe how as a life-form the tabby cat lives over the course of a life. Further, while examples of Aristotelian categoricals can be lax at times, a species or life-form that is the subject of an Aristotelian categorical picks out a very specific species, such as tuna, and should not be thought of as picking out the wider genus fish.
Aristotelian categoricals are further characterized by their irreducibility to other logical forms. Thompson argues that Aristotelian categoricals are not reducible to statistical propositions, propositions about most members of a species, universal propositions about a species, generic propositions, propositions with ceteris paribus clauses, nor a universal normative proposition about a species. Perhaps the easiest mistake to make in trying to understand what Aristotelian categoricals are is to think of them as statistical propositions. To do so is to think that Aristotelian categoricals express the activities the average member of a species does, or to think that if most members of a species are a certain way or perform certain activities then there is an Aristotelian categorical that represents this statistical norm.
For instance, if the majority of cats observed are three legged, then it might be thought that there is an Aristotelian categorical such as“The domestic cat is three legged”. Thompson does not think that Aristotelian categoricals are given by such observation of the majority of cats, and might say that we would think that all the three legged cats are missing a leg, rather than infer from observations of three legged cats that “The domestic cat is three legged” is an Aristotelian categorical,. As such, there can be Aristotelian categoricals for a species that the majority of the species does not meet.
Another reason Thompson does not think Aristotelian categoricals give statistical norms is because he thinks “…just as “All A’s are F” and “All A’s are G” together entail “All A’s are both F and G”, so also “The S is F” (or “S’s are F”) and “The S is G” (or “S’s are G”) together entail “The S is both F and G” (or “S’s are both F and G”) if it is our sort of combination that is expressed. The inference would obviously be invalid for any sort of statistical generalization.” To clarify the inference’s obvious invalidity, suppose that a species of dog is composed of exactly one hundred members. Further, suppose that fifty one percent are black, and fifty one percent have long hair. If statistical generalization yielded Aristotelian categoricals, then we might say that “Dogs are black” and “Dogs have long hair”. Combining the categoricals, we get that “Dogs are black and have long hair”. If Aristotelian categoricals are statistical generalizations, this combining would entail that approximately fifty percent or more dogs are black and have long hair.
While most dogs being black and most dogs having long hair may be truths, that less than fifty percent of dogs be black and have long hair is still possible. For if fifty one dogs are black and fifty one dogs have long hair, it might be that only five dogs are both black and long haired; forty six black dogs have short hair, forty six long haired dogs are not black, and three dogs are neither black nor long haired. So such an inference is invalid if Aristotelian categoricals are statistical generalizations, and Aristotelian categoricals can be combined in Thompson’s manner. Thus that Aristotelian categoricals are statistical generalizations is false, given that that Aristotelian categoricals can be legitimately combined.
Finally, to further make clear why the inference is obviously invalid for any sort of statistical generalization, we only need to reflect “…that by repeated application of our apparently unexciting rule of inference “S’s are F”, “S’s are G”, ergo “S’s are both F and G” we will presumably always be able to produce a true statement of our form involving a complex conjunctive predicate that is not true of any member of the kind denoted by its subject, living or dead. I mean: nobody is perfect.”
Aristotelian categoricals’ grammatical and logical form
Having characterized somewhat the kind of thought the sentences called ‘Aristotelian categoricals’ are meant to pick out, I now turn to the grammar and logic of such sentences. Looking at Aristotelian categoricals in more detail, we might wonder what can be said about the grammatical or logical form they take. Can a preferred form of grammar isolate all Aristotelian categoricals? Is there even a preferred form of grammar? Examples of proposed Aristotelian categoricals suggest that there is such a preferred form. In Thompson and Foot’s writing, the following are proposed as Aristotelian categoricals: ‘The blue tit has a round blue patch on its head.’ , ‘The deer is an animal whose form of defense is flight’ , ‘The male peacock has a brightly colored tail’ , ‘The medusa of the umbrella jelly has one hundred and forty four tentacles’ , ‘Upon fertilization, the mature umbrella jelly lays hundreds of eggs’ , and ‘The domestic cat has four legs’.
Generalizing from these examples, we might say something more about a preferred form. As observed by Thomson, Aristotelian categoricals canonically take the form “The S is/has/does F”, but can also take other forms such as ‘S’s are/have/do P’ or ‘S’s typically are/have/do P’. Thompson writes that it is necessary that a common noun stand for the ‘S’ and some other predicative expression ‘P’ be present or in the offing. Since Aristotelian categoricals are about life-forms, it is perhaps helpful to think of the ‘S’ ranging over species or kinds.
Thompson elaborates more on the preferred form of Aristotelian categoricals in his paper “Apprehending the Human Form”. Thompson writes:
“We may call the judgments you are assembling in your
monograph natural historical judgments of type B. Their
general form will be something like this: the S is/has/does F,
or S’s are/do/have F, or S’s characteristically are/have/do F,
or it belongs to an S to be/do/have F, or this is (part of) how S’s
live: they are/do/have F. What particular verbal materials are
used to join ‘S’ and ‘F’ in speech is of no importance; what
matters is that the resulting nexus of signs, perhaps taken together
with features of the context, expresses a distinctive form of judgment.”
Michael Thompson, Apprehending the Human Form, pg. 2.
Thompson also thinks that Aristotelian categoricals have unusual and distinctive temporal properties. Thompson writes:
“Of any individual jelly here and now, you will speak in the
usual temporal way. You will judge that it ‘is’ in some one of
these phases and ‘has been’ in another, and you might form the
expectation that it ‘will later on be’ in another. But of the umbrella
jelly as a general kind, or form, of life, you will speak in the first
instance completely atemporally. In your monograph about the life
form, you will say that on its first appearance the thing ‘is’ an egg,
then later it ‘is’ a polyp, then later it ‘is’ a medusa. Or again: of this
umbrella jelly hic et nunc, you will say that it ‘is developing’ into a
medusa; of ‘the’ umbrella jelly you will say that it ‘develops’ into a
medusa.”
Michael Thompson, Apprehending the Human Form, pg. 2.
Having characterized the general form of Aristotelian categoricals, further distinctions can be made. Among the propositions that have the correct form of Aristotelian categoricals, there will be some that are true and some that are false. By replacing the predicate in an Aristotelian categorical, a true Aristotelian categorical might be made false. For example, supposing “the penguin breathes air” and “the anchovy breathes through water” are true Aristotelian categoricals, changing the predicate around yields false propositions that are presumably still Aristotelian categoricals. As an Aristotelian categorical “the penguin breathes through water” is false, although still of the same logical kind as true Aristotelian categoricals. This is significant, as some replacements of the predicate appear to yield propositions that are grammatically like Aristotelian categoricals, yet do not seem to fall under the same logical kind. For example, the sentences “the penguin is my favorite animal”, or “the penguin is cute” are grammatically like Aristotelian categoricals, but do not appear to have the right content.
For clarity, I will call propositions that are about species’ natural history or vital descriptions ‘life-form propositions’. This category called ‘life-form propositions’ is intended to capture the logical kind given by Aristotelian categoricals. Life-form proposition includes all propositions that are about a species’ natural history, both true and false. The sentence “the penguin is my favorite animal” can thus be said to have the preferred grammatical form of an Aristotelian categorical, but is nevertheless not a life-form proposition. This category helps in sorting propositions, for given a proposition, we might ask if it even is a life-form proposition, let alone a true or false one.
We now might ask whether every sentence of the preferred grammatical form is a life-form proposition. A tentative answer to this is no, as some sentences with the preferred form do not appear to be life-form propositions. As mentioned already, propositions such as “the penguin is cute”, “the puffer fish is scary”, and “the black bear is my friend’s favorite animal” does not appear to be a life-form proposition, although it apparently shares the same form as the former sentence. So the preferred grammatical form is insufficient for picking out life-form propositions.
One criticism with this conclusion is that the examples given may confuse the actual subject and predicate of the sentence. For the proposition “the black bear is my friend’s favorite animal”, we might treat ‘my friend’ as the subject of the sentence, rather than the species black bear. So, sentences like “the black bear is my friend’s favorite animal” might be read as “my friend’s favorite animal is the black bear”. Read this way, the sentence does not express a predicate of a species, but rather of an individual. The sentence will thus not have the form of an Aristotelian categorical, which are about species or kinds, but of a proposition of first order logic. This method also seems to apply to other predicates such as ‘is believed to be’, ‘is suspected to be’, ‘is hated by’ and other sentences that might express a subject’s psychological attitude towards a species.
Another reason to think that the preferred grammatical form does not isolate all life form propositions is that some sentences of the preferred grammatical form appear ambiguous between being life form propositions and more general propositions. Consider that a certain species of tumbleweed might dry up in the late spring and tumble down hills in order to spread its seed, such that in talking about the life form it is true that ‘tumbleweeds tumble down the hills in the late spring’. We should easily admit that this proposition is a life form proposition.
Now consider a situation in which, in the late spring, many goat-snares are set up in the hills, bated with goat food, such that when caught, goats tumble down the hill. If this is a common event, a local might shrug when asked by a wide-eyed tourist what is happening and say “goats tumble down the hills in the late spring”. In this situation, what the local means by his statement is clearly not a life form proposition, although his grammar is the same. While we could imagine a situation in which the local is mistaken about the natural history of goats, this is not happening here. The local knows all about the snares, and does not think goats are supposed to tumble down hills in the spring. This situation shows that some sentences of the preferred grammatical form are ambiguous between life-form propositions and general propositions.
How might this ambiguity be eliminated? Is there a clear way to separate the life form proposition in this sentence from the more general proposition? Of a similar case Thomson writes “But background knowledge, my alarmed tone of voice, and the predicate I use in saying “The domestic cat has three legs” will show that here I am not making an attempt at natural history, that it is poor Tibbles that I am talking about, and that my statement has the ‘logical form’ of “The cat is on the mat”. Thomson’s method here seems a little weak. How the predicate he uses in saying “the domestic cat has three legs” helps to disambiguate the two logical forms is unclear. In his later paper, Apprehending the Human Form, Thomson writes “What particular verbal materials are used to join ‘S’ and ‘F’ in speech is of no importance…” so there seems to be no restriction on what can be predicated. Thus merely having the predicate ‘has three legs’ does not disambiguate the two forms, nor presumably other predicates. Further, Thomson is not specific on what background knowledge he is appealing to, and some cases of ambiguity might be absent of alarmed tones of voices, let alone voices at all in the case of writing.
Thomson seems to have a stronger reply available based on his observations about Aristotelian categoricals’ tense. Recall that Thomson writes “But of the umbrella jelly as a general kind, or form, of life, you will speak in the first instance completely atemporally.” A method for disambiguating propositions that are both life form propositions and general form can be based on this observation. Since life form propositions are atemporal, we can separate the general sense from an ambiguous proposition by changing/re-describing its tense.
For example, the general sense of the proposition “the domestic cat has three legs” will be able to be tensed to “Three minutes from now, the domestic cat will have had three legs”, while in the life form proposition will not. Likewise, for the general proposition “Goats tumble down the mountains in the late spring”, we can meaningfully change the tense, while for the life form proposition such a change is not possible. For the general proposition, it follows that if “Goats tumble down the mountains in the late spring”, then “In the summer, goats will have tumbled down the mountains in the late spring”, whereas this is not a consequence of the life form proposition.
This method also seems helpful towards some of our previously considered propositions. For the truth of “the black bear is my friend’s favorite animal” will be temporally sensitive; if my friend changes her favorite animal, or dies, then we are inclined to think that the proposition becomes false, whereas life form propositions aren’t temporally sensitive in such a manner. Likewise, we say of propositions such as “the puffer-fish is scary”, that if the truth-value of such a sentence is sensitive to tense, then we are considering the general proposition, and not the life form proposition.
This method is not entirely without flaws. One immediate worry with it is to the status of extinct species. If life-form propositions exist about extinct species, we might think that the tense would change. For example, in talking about dinosaurs, we say things such as “the T-rex was carnivorous”, rather than “the T-rex is carnivorous”. The status of such sentences as life-form propositions matters especially when connection are drawn from the relation between an individual and its species. A further worry is that Thompson might be wrong about life-form propositions being atemporal. For sentence such as “the black bear, having gorged itself in the spring, hibernates in the fall” are tensed and do appear to be life-form propositions.
The considerations suggest that life-form propositions cannot be isolated by grammatical form alone, and thus a method of representing them logically is desirable. This is not a suggestion that Foot and Thomson are likely to find objectionable, for Foot at least thinks of life-form propositions as of a different logical kind than general propositions, and would probably be open to representing its logical form in a distinctive manner. A provisional suggestion is that life-form propositions be represented in a modal manner, with the operator ‘ c’ modifying the truth-value of a sentence it is applied to. Some motivation for considering using an operator that modifies the truth value of a sentence comes from the grammar of certain examples given by Thomson, such as “…this is (part of) how S’s live: they are/do/have F.” and “…’This is how things go with an S: it does F’.” In these sentences, the sentence preceding the colon appears to modify how we read the sentence following the colon. As such, we might perhaps think of the preceding sentence as an operator that applies to the following, which again we will represent with the symbol ‘c’. What the semantic interpretation of ‘c’ is, and whether we will need to represent ‘c’ differently for each species will be approached through considering the way life-form propositions interact with other propositions, and the different inferences we might make from these pairs.
How Do Life-Form Propositions Interact With Other Propositions?
Having given preliminary reasons to think there are such thing as life-form propositions, and having characterized something of its grammatical and logical nature, we will now examine how life-form propositions interact with other propositions. Of particular importance will be how true life-form propositions interact with facts about individuals of the life-form. The following ‘schemas’ propose rules of inference between life-form propositions, and general propositions about individuals of that life form. These schemas may be thought of as somewhat analogous to Kripke’s example of the relation between two actual dice and the various possible states of the dice. As we can conclude that there is a one in thirty-six chance that the actual two dice will both role a three, based on how the actual dice are related to possible states, so we might conclude that an actual individual wolf’s broken leg is an accident, or a defect, based on how the wolf is related to its species.
More specifically, these schemas will look at a sort of mismatch between the general life-form proposition, and the particular propositions about some life form member. For example, we will consider how to think of the mismatch between the life-form proposition “The domestic cat is four legged”, and propositions about particular cats, say Tibbles, who is three legged. How should this mismatch be characterized?
Accident Schema
Interaction between life-form propositions and general propositions appears to provide a notion of accident. While a schema entailing accident is not proposed by Thompson or Foot explicitly, it is drawn from statements by Thompson such as: “What we miss, or miss most obviously, in the flat physicalistic picture of reproduction, is any conception of the unity of a thing and its circumstances as potentially non-accidental. An organism’s coming to be in such circumstances as tend to its reproduction is itself typically a vital operation, a phase in a life-process, and therefore, in a certain sense, ‘no-accident’.” As Thompson has observed, there is a sense in which “…the unity of a thing and its circumstances…” is non-accidental. I think Thompson should additionally hold that there is a sense in which facts about an organism are accidental under a life-form description, given that he already accepts a non-accidental sense. I will now argue that one sense of accident that Thompson ought to accept occurs for all mismatch between life-form and organism.
What is an accident? Philosophers often talk of properties being accidental, by which they mean something like contingent and identical. That Socrates drank Hemlock, and not Nightshade, may be thought a contingent feature of Socrates’ identity, for Socrates would still be Socrates if he did not drink Hemlock. As such, the property of being a hemlock drinker can be said to be an accidental property of Socrates. I am not primarily interested in this sense of accident. I will discuss why this is more in the section entitled modality.
Another way accident is used is to pick out a coincidental feature, or to explain a relation through chance. Using a notion of accident like this, we might say that a frog’s being born during the president’s speech is an accident. Likewise, that Tom met Sally on her way from the market may also be an accident in that the meeting was a coincidence. I am somewhat interested in this use of accident, although I am unsure if the sense of accident that mismatch between life-form and organism gives this sense or not. I will discuss the possible connection of this sense with the sense I am interested in the section entitled coincidence and chance.
The sense of accident that a mismatch gives is that of something not supposed to happen happening, or of something supposed to happen not happening. I take car accidents to be an example of this sense. When a car accident occurs, something not supposed to happen, the crash, does happen. Likewise, we might say a car accident occurs when something supposed to happen, the non-crashing, does not happen.
Now, this sense of accident occurs for all mismatch of “supposed to” claims, independent of the grounds of why something is “supposed to” occur. As such, I do not think the explanation of why something is “supposed to” occur will always be due to a life-form proposition. For example, we might say that a man who intends to hit the eight ball in the corner pocket is supposed to hit the eight ball in the corner pocket, given his intentions. If the man then hits the eight ball into the side pocket, we might say that the mismatch of what was supposed to happen and what happened is an accident. Here though, what is “supposed to” happen is determined in virtue of the man’s intentions. So all things “supposed to be” are not “supposed to be” in virtue of some life-form proposition. On the other hand, I think all life-form propositions determine what is “supposed to be” for some life-form. For given some life-form proposition, we can say that organisms of that life-form are “supposed to be” as the life-form proposition holds. So all life-form propositions determine what is “supposed to be”.
Importantly, what is “supposed to be” in virtue of life-form is species dependent. For facts about the life-forms cougar and deer might make it the case that both this cougar is supposed to catch the fleeing deer, and that this deer is supposed to escape the cougar. As such, opposing ‘supposed’s are not contradictory if they are of different species. The same can be said to apply to different stages in a life; things that are supposed to occur at a certain stage in a life might not be supposed to occur at other stages. While infant horses are supposed to drink its mother’s milk for, we do not think that adult horses are supposed to drink its mother’s milk. As such, in representing life-form propositions, what life-form a life-form proposition applies to ought to be kept track of. Further, the sense of accident given by a mismatch of life-form proposition and general proposition will be life-form relative. So if the cougar catches the deer, it is no accident in light of the cougar’s life form that the deer is caught, but is an accident in light of the deer’s life-form that the deer is caught. Vice versa if the deer gets away.
I think we make judgments using this sense of accident throughout our lives. For example, when a man has one arm, and a child points to a man and asks “Mommy, what’s wrong with him?”, the child appears to recognize the man’s being one armed as an accident. In this case, the child’s use of “wrong” appears to point out that there is a discrepancy between two like objects, like when given two similar pictures with the question “What’s wrong in this drawing”. Some objects are missing in one picture, such as where in one picture a bucket is striped, and in the other it is not. Further, this feeling of wrongness seems to be a sign that the thing thought to be wrong is an accident of some sort.
The child’s belief that the man is “supposed to” have two arms is interesting, for the child expects that the man is supposed to have two arms while recognizing at the same time that the man has a single arm. This kind of expectation is rather odd, for a person who expects this expects something to be the case that she knows is not the case. Further, there seems to be a difference between the child’s expectation that an arm is supposed to be there when it is not, and say, the expectations of a man who is amazed that a ball, when let go, falls down rather than remains where it is. Suppose such a man made a similar claim: the ball is supposed to stay in place, not fall down. We might think him slightly mad. Yet both he and the child think the world is supposed to be in a way that the world is not, but we think the child’s supposition reasonable and the man’s not. So the difference in reasonability is not due to a supposition that the world be other than it is, but rather, in one case we do think that the man is supposed to have two arms, and in the other, that the man is wrong that the ball is supposed to remain where it is.
Now, I don’t think the answer “He’s missing an arm” is really responsive to the child’s inquiry. For saying that “he’s missing an arm” seems just to recognize that the man is supposed to have two arms, and only has one arm. And this just is the discrepancy the child is pointing to. Rather, the child might just as naturally have asked “what happened to him” upon noticing the man with one arm. So what the child is asking for is an explanation of how the man became one armed when the kind of thing he is is two armed. Further, we think such explanations exist. If we ask the man why he only has one arm, we expect there to be some historical fact about how he lost his arm. In the case of the madman however, we don’t think there is an explanation how the ball came to fall when the kind of thing it is remains in place, because we don’t think the kind of thing it is actually remains in place. We think balls are supposed to fall, due to the laws of gravity.
Interference Principle
When something is not as it is supposed to be, we say that it is by accident that it is thus. If there are facts about what events or states are supposed to obtain for a life-form, and it is also true that those facts don’t obtain, then we expect there to be an explanation why. This explanation determines some kind of interference that makes what is supposed to be not occur. In the missing arm case, if a person thinks an arm is supposed to be in a certain case, and the arm is not there now, he will reasonably expect there to be an explanation for why the arm is not where it is supposed to be. He might ask if the arm was cut off, or if a disease inhibited its growth, or other such questions. These questions are about the empirical history of the arm, and are questions that could be potentially verified or rejected empirically.
So, in cases of accident characterized by a life-form mismatch, the accident can be explained by some interference upon a creature’s activities such that the individual is unable to match how it is supposed to be. For instance, a duck might not fly, although it is supposed to, due to a broken wing. In such cases, accident is linked to some kind of interference on an organism. If all cases of accident are linked to interference, then we could hold that for any claim that something is wrong with a life form, there will be an explanation for why something is wrong with it in terms of some interference.
Are all cases of accident linked to interference then? Suppose not. Then, there would be some accident occurring without interference. This means that there would be a true proposition about how an individual is supposed to be that the individual does not meet, but nothing would explain how the individual is not as how he is supposed to be. An example of such case is when a plane leaves on time, it goes the correct speed to reach its destination on time, there was no wind, the motors don’t fail, its not hit by a meteor etc, and the plane is supposed to be at the airport at a certain time, yet the plane still is not at the airport at the appropriate time. I am tempted to say that such cases should be called miracles, yet even then an explanation can be given- some deity or supernatural event caused the case. So I am reluctant even to classify the case here.
While I thus lack a formal proof of the principle that all cases of accident involve some kind of interference, it seems intuitively plausible, and appears to be implicit in some of the literature regarding life-form principles. For example, when Thompson writes of an imagined biologist’s experience of discovering an imagined new species, he writes: “The specimen in question at first strikes you as a bit of a freak, perhaps. You wonder whether it might not be a defective instance of a species familiar to you, perhaps the cross jelly. You consider that maybe its development has been compromised by some complex chemical pollutant abroad in these water….” Here, Thompson’s biologist immediately goes from the thought that the specimen is defective, that something is wrong with it, to look for some explanation of an interference in the cross jelly’s life, which in this case would be some complex chemical pollutant in the waters. As such, I tentatively propose the following principle about accidents and interference: for any case of accidence, there will be an explanation in terms of an interference why the species member in question is not as it is supposed to be. This principle will be called ‘the interference principle’.
Modality
Whether the sense of accidence given by life-form propositions can be integrated with contemporary modal views of accidence or essence is natural to wonder, and an interesting question. Here, I will argue that the sense of accidence outlined above does not reduce to contemporary modal notion, nor seems to be related immediately to a contemporary notion of essence.
The concept of accidence outlined does not appear to be modal, for if something is supposed to be, it does not follow that it must be. For, if one claimed that “the domestic cat is supposed to be four legged”, the fact that a cat is not four legged does not contradict the ‘supposed to be’ fact. If ‘supposed to be’ is interpreted as ‘necessarily’, such that “the domestic cat is necessarily four legged’, then a domestic cat’s three leggedness would contradict the claim’s necessity. Further, if it is the case that the domestic cat necessarily is four legged, then no accidents involving three legged cats are possible, since there will never be any incongruity between how the cat is and how it is supposed to be. Since life-form propositions can fail to match how a member of the life-form is, necessity gives too strong of a reading to the “supposed to”. So “supposed to be” in virtue of life-form is not equivalent to necessity, and as such, the notion of accident given by mismatch of these seems distinct from a notion of accident as a non necessary, or contingent, feature.
Further, the concept of accidence outlined does not appear to be a species of the contemporary notion of essence and accident. What is “supposed to be” is not equivalent to a notion of essence, either thought of in modal terms or as a real definition. If what is “supposed to be” were thought of as necessity de re, then it appears that we would have the same contradiction as outlined above, as “supposed to be” facts would prevent cases of what we are calling accidence. And if what is “supposed to be” is treated as a real definition, modeled on Kit Fine’s work, we have a problem too, for Fine writes in his paper “Essence and Modality”:
“Certainly, there is a connection between the two concepts
[necessity and essence]. For any essentialist attribution
will give rise to a necessary truth; if certain objects are essentially
related then it is necessarily true that the objects be so related
(or necessarily true given that the objects exist). However,
the resulting truth is not necessary simpliciter. For it is true
in virtue of the identity of the objects in question; the necessity
has its source in those objects which are the subject of the underlying
essentialist claim.”
Kit Fine, “Essence and Modality”, pg. 7
So, under Fine’s conception of essence, if ‘supposed to be’ is thought as ‘essentially’, as in “cats are essentially four legged”, then it follows that it is necessary that the objects be so related. Thus, cats being necessarily four-legged would follow, and this result has already been shown to be at odds with the account of accident developed. For these reasons, I am not primarily interested in the modal sense of accident.
Coincidence and Chance
The use of accident in picking out a coincidental feature, or to explain a relation through chance is interesting because whether Thompson’s use of accident is of this sense, or that of a mismatch is unclear. Thompson writes of accident that:
“An organism’s coming to be in such circumstances as tend to its
reproduction is itself typically a vital operation, a phase in a life-process,
and therefore, in a certain sense, ‘no accident’. A dandelion-seed’s falling
on reproductively apt soil may seem fortuitous, but its parent, by a kind
of ruse of vegetation, makes such an accident no accident, most obviously
by producing so much seed.”
Michael Thompson, “The Representation of Life, pg 21
Thompson’s use of accident here seems to hold that vital operations make an act no accident in the sense that the act is not by chance, or not a coincidence. For the description of the seed’s falling being fortuitous makes the seed’s falling appear a beneficial chance event, and the negation of this seems like a non-coincidental event. Thompson’s explanation of this seems strange though, for he holds the act was not a coincidence because the flower produced a lot of seed. This makes his explanation of the non-accidence of the event seem like an appeal to favorable statistics, which he later argues is not the kind of norm Aristotelian categoricals give. Nevertheless, Thompson seems to have in mind a different sense of accident than I argue for.
Now, I do not think this is a problem for the account I give, as I think Thompson should accept the sense of accident I propose. Yet Thompson’s claim does seem to hint at a more interesting notion of accident in that his use makes events that are only by chance given a physicalistic picture not by chance when life-form enters the picture. So it would be nice if I could relate the notions of accident such that I capture Thompson’s claim. I am not sure if this can be done, but I would like to make some observations about coincidence and chance that seem to hint at some relation between senses.
First, I would like to question what a coincidence is, or when something occurs by chance. I will focus more on coincidence. I will take what people commonly say about coincidence as a starting point. Wikipedia can surely be said to express common opinions said about some topic. The Wikipedia entry on coincidence writes: “Coincidence is the noteworthy alignment of two or more events or circumstances without obvious causal connection. The word is derived from the Latin co- (“in,” “with,” “together”) and incidere (“to fall on”).”
This characterization of coincidence can be refined in light of examples of coincidences. My first observation is that the “noteworthy alignment” of events need not mean that the events happened the same time, nor the same place. For example, say Roosevelt corresponded with Japan in a private phone call on December 6th. Further, Japan bombed Pearl Harbor on December 7th. These events don’t happen at the same time or place, but of the conjunction of the two we say “That Roosevelt corresponded with Japan on December 6th and Japan bombed Pearl Harbor on December 7th is a coincidence.” So events do not need to occur at the same time or place.
Further, coincidence seems to be said of some conjunction. Further, while in some cases objects conjoined in a coincidence claim can themselves be coincidences, the same coincidence is not said of each of the objects conjoined, but only of the conjunction. For example, suppose Tom strolled around and entered the bar. That Tom strolled around and entered the bar can be a coincidence. Further, suppose Sue strolled around and entered this bar. That Sue strolled around and entered this bar can also be a coincidence. But these coincidences are different than that said of Tom and Sue both strolled around and entered the bar. In the former cases, the lack of connection is between the single subject strolling around and entering the bar, and in the latter the lack of connection is between the pair strolling around and entering. Further, both Tom and Sue’s strolling around and entering this bar might not be a coincidence, if perhaps they both planned on entering the bar after strolling, yet their both entering this bar may still be a coincidence, if they did not plan on both of them entering the bar after strolling. So coincidence appears to be said of a conjunction, and not a conjunction of two coincidences, but any conjunction of two things.
The objects of a coincidence are unclear, and events are insufficient as objects. My name is Michael. A character in the show “Lost” is also Michael. Neither of these are events, yet that “My name is Michael and a character in “Lost” is Michael is a coincidence.” This seems to suggest that the objects of a coincidence are not what determines the conjunction of objects to be a coincidence, but rather, the relation between objects. Further, while the clause “without obvious causal connection” describes the relation missing in common use, or even perhaps “true” instances of coincidence, why other relations missing aren’t sorts of coincidence is not clear to me. That Jennifer’s middle name is Shasta, and a mountain is named Shasta is not a coincidence, for her parents intended her middle name to be that of the mountain. Yet that Jennifer’s middle name is Shasta, and line of stores is named Shasta is a coincidence, for neither Jennifer’s parents nor the store line intended for the names to be the same. Neither relation has an obvious causal connection, but rather, a relation depending on intentions.
If such reflection is correct, then different kinds of coincidences might exist, the difference of coincidence depending on the kind of relation lacking. With “laws of nature” as the relevant relation, we would say that some cue ball hitting a pair of balls and the two balls going into the corner pockets is not a coincidence, for the laws of physics relate the events in the right way. We may even say that by the laws of physics, the two balls are supposed to go into the corner pockets given their being hit by the cue ball in that manner. To not do so would be to disobey the laws of physics. So this relation holding is no coincidence, and never could be, given laws of nature.
That an event cannot be a coincidence given relations of laws of nature does not mean that the event is no coincidence given some other relation. If there is a relation between intended events and actual events, we seem to get a kind of coincidence other than that given by laws of nature. Suppose of the same event that Tom, the pool player, does not intend for the two balls to go into corner pockets. That Tom hits the cue ball such that the cue ball hits the pair of balls and the two balls go into the corner pockets is a coincidence. The event as it happened has no relation to Tom’s intentions, either because he had no intentions, or he intended something that did not happen. The relation between the cue ball’s hitting and the ball’s going into the pocket is no coincidence under laws of physics, but the relation between the cue ball’s hitting and the ball’s going into the pocket is a coincidence under a description of intentions.
Like the case of intentions, life-form propositions appear to allow the “…conception of the unity of a thing and its circumstances as potentially non-accidental.” Thompson explanation of why a seed’s striking reproductively apt soil is not fortuitous seems to be that there is a relation between this event and the life-form. The description of unity of a thing and its circumstances is somewhat ambiguous however, for it masks what the unity is between. Is the unity between some thing and some circumstances, or between some thing and that thing’s circumstances? Unity between the former does not appear to make use of the Aristotelian categoricals Thompson is arguing for and so seems somewhat uninteresting. If the latter is the claim, then what is the first thing? I think Thompson should say life-form, and that the unity is between life-form and member of that life-form’s circumstances.
If Thompson accepts this, then he can account for the non-coincidence of a flower’s seed striking fertile ground in a manner like that of the pool player. That a flower’s seed strike fertile grounds may be a coincidence, or chance given physical description, for it is only by statistically lucky circumstances that they do this. At the same time, the event is not a coincidence, for given the flower’s life-form, this event is supposed to happen, not because of anything’s intentions, but because that’s what flowers do in these circumstances if they match the life-form. So, a flower’s seed striking fertile ground in some circumstances is no coincidence, for such an event is correctly related to the organisms life-form.
Thompson does not say that the unity is between a life-form and a member of that life-form’s circumstances however. Thompson’s answer is that the production of so much seed by the seed’s parent explains why the striking fertile ground is not an accident. What explains the event being non-accidental in this example seems to be a statistical fact. Given that the flower produced so much seed, some seed was bound to strike fertile ground. Perhaps Aristotelian categoricals are meant to explain why the flower produces so much seed. If so, then the connection is not clearly laid out here.
Thompson’s answer is further complicated considering that Thompson uses the term “a dandelion seed’s” and not “some dandelion seed”, for we still might think that it is fortuitous that this seed struck fertile soil, while many other seeds did not, even when the seed’s parent produced so much seed. Further, this example of chance seem present even given a life-form, for presumably it’s part of the dandelion’s life-form that some seed strike fertile ground, and not that any seed in particular strike ground. So it is odd to think that the fortuitous event results from this seed striking fertile ground, as the unity of a thing and its circumstances does not seem to remove the chance from the event.
Interpreting the “c” Operator
Given our reflection on understanding the mismatch between life-form proposition and general proposition as a kind of accidence, I thus propose using “supposed to be” as a semantic interpretation of our earlier c operator. Further, as we have reflected that “supposed to be” is relative to some species, we ought to keep track of the species. This can be done through subscript. The accident schema might be represented as follows: A proposition about an individual member of a species S is said to be accidental when there is something wrong with the individual S or an event involving an individual S. Further, something is wrong with an individual S or an event involving an individual S if and only if:
1w. It isn’t as it supposed to be, or
2w. It is as it is not supposed to be.
In terms of pairs of true facts about a life form and true facts about an individual, the following are conditions for satisfaction:
1w. cP "x (P(x)’Q(x)), P(a)&~Q(a), or
2w. cP "x (P(x)’~Q(x)), P(a)&Q(a).
Likewise, nothing is wrong with being Q, or equivalently, things are right or non-accidental with an individual S or event involving an individual S if and only if:
1r. It is as it is supposed to be, or
2r. It isn’t as it is not supposed to be.
Again, in terms of pairs of true facts about a life form and true facts about an individual, these are satisfied if and only if it is the case that:
1r. cP "x (P(x)’Q(x)), P(a)&Q(a), or
2r. cP "x (P(x)’~Q(x)), P(a)&~Q(a).
These various forms of satisfaction all follow from the basic way in which accident is conceived, which can be represented as A(F(a)&S(a))´ (cF "x (F(x)’~S(x) ) & F(a)&S(a)). If we allow for substitution, we get:
1. A(F(a)&S(a))´ (cF"x (F(x)’~S(x)) & F(a)&S(a))
2. A(F(a)&~S(a))´ (cF"x (F(x)’S(x)) &F(a)&~ S(a))
Finally, the accident schema is asymmetrical in that only accidents require an explanation. The logical equivalences to this schema will thus not fall under the interference principle, and not require an explanation.
Further Relation?
If we assume ~A(R(a)&S(a))’ (cR "x (R(x) ’S(x)) & R(a)&S(a)), as well as our accident schema and the principle of interference, we get an obviously too strong principle, for from any event being non accidental, we could infer that such an event is part of the organism’s life-form’s kind. For if some cat’s being on the mat is no accident, then “Cats are on mats” becomes a true life form proposition by modus ponens. This relation is thus much too strong of an assumption. However, we still might be tempted to assume (cR"x (R(x)’S(x))&R(a)&S(a))’ ~A(R(a)&S(a)). This rule states that if there is a match between how an organism is and how its life-form is, then how the organism is is no accident. This principle has interesting consequences about the use of technology or surgery in correcting defect, and it is not clear whether the principle is obviously false, or if it is true. I will discuss this principle more in regards to my discussion of Foot.
The Evaluative Schema
In claiming that the mismatch between life-form propositions and general propositions should be understood as a kind of accidence, my account of life-form propositions is put somewhat at odds with Thompson’s account, for Thompson also proposes a way of understanding the mismatch between propositions that varies from my own account. Thompson proposes a rule of inference between life-form propositions and general propositions about an individual member of a life-form. I will call this rule the evaluative schema for clarity.
Thompson suggests that from an Aristotelian categorical ‘S’s do P’, and a proposition about some specific member of S, such that ‘This member of S, does not P’, it is valid to infer that ‘This member of S is defective in that it does not P’. An example of such an inference is, from the life-form proposition that “ The domestic cat is four legged”, and a proposition about Tibbles such that Tibbles the cat is three legged, the conclusion that Tibbles is defective in that he is three legged can be inferred. Of the same propositions, my account would conclude that there is something wrong with Tibbles being three legged, as he is not as he is supposed to be, so Tibbles’ three leggedness is an accident.
My account thus far is weaker than Thompson’s in that the conclusions it draws are not yet at the level of defect; my account is neutral as to whether the accidence is detrimental to the organism. The difference between accounts being noted, it will now be ignored in favor of discussing Thompson’s evaluative schema. As will be seen, the difference in accounts will eventually be useful in reconciling discrepancies between Thompson and Foot’s uses of Aristotelian categoricals that arise naturally from considering Thompson’s evaluative schema.
Finally, the evaluative schema holds that a sufficient condition of being defective is to be a member of a species that does not fulfill some predicate of that species. Defect is also species relative, a consequence Philippa Foot appears to think is desirable when she writes “Thus, evaluation of an individual living thing in its own right, with no reference to our interests or desires, is possible where there is intersection of two types of propositions…” and “But features of plants and animals have what one might call an ‘autonomous’, ‘intrinsic’, or as I shall say ‘natural’ goodness and defect that may have nothing to do with the needs or wants of the members of any other species of living thing…”
The Converse Schema
One might strengthen the connection between defect and Aristotelian categoricals by holding that the relation between being defective and lacking a certain parity between Aristotelian categoricals and propositions about a member of a species holds conversely as well. This would be to treat having a disparity between Aristotelian categoricals and facts about individuals as a necessary condition for defect. If one held that the converse of the evaluative schema is a necessary condition for defect, then one would hold that the propositions ‘This member of S is defective in that it does not P’ implies that there exists a life-form proposition such that ‘S’s do P’ and a general proposition that ‘This member of S does not P.’
While Thompson does not consider the converse rule, we can still examine reasons why or why not the converse rule should be adopted. Adopting the converse of the evaluative schema in addition to the evaluative schema would provide a complete account of defect, providing both necessary and sufficient conditions for judging a thing defective. Further, adopting the converse rule would provide the basis for arguing that certain species exists and that certain life-form propositions are true.
The intuition that something is defective would provide the basis for arguing that a species exist because, if all defects are a matter of an individual not bearing a certain relation to its species, then in any true case of defect there will fail to be a certain relation between an individual and its species. Further, an individual cannot fail to bear a relation to its species if there is no species it is of, for failing to bear a relation requires that a relation could be born. So, if the converse rule holds for all defects, then in any case where something is truly defective, the converse rule would suggest that the defective thing is a member of some life-form.
The converse rule would provide the basis for arguing that certain life-form propositions are true because, in any true case of defect, there will be a pair of true propositions, one of which will be a true proposition about how an individual is, and the other a true life-form proposition about the species the defective member is of. The life-form proposition will be such that the fact about the individual is at odds with the life form proposition which as species that a member fails, and a species that the member is of. Intuitions about something being defective, and being able to explicate what in particular is wrong with the individual can thus serve as a guide to stating how the species is supposed to be.
The converse rule appears too strong in that it commits to some things having a species that seem intuitively not to have a species. This objection is especially forceful since the ‘species’ concept Thompson talks of appears to only apply to living things, for this consequence holds that when certain things thought to be non-living can be thought of as defective, this entails that that these non-living things have a species. For example, when we sometimes talk of different devices being defective, like when a television has a broken part, we appear to truly attribute to the television a defect, yet we do not take the television to have a life-form. If one were to insist on the converse rule, then the concept of species and life-form associated with it would be much more broadly applicable than it formerly appeared, and would perhaps not pick out living species per say, but perhaps just species.
The converse rule might be toned down to meet such an objection. A more moderate converse rule can be considered that holds for certain defects, namely, defects in an individual as a life-form. While a full account of defect would not be given, and a way of telling when a defect is species specific is needed, this would still allow for the useful consequences of the converse rule to be employed. A method of determining if the converse holds is to see if an account of defect not relying on the evaluative schema can be made. If such an account exists, then there is reason to think the converse rule is not necessary in explaining all forms of defect.
Attribution
Mismatch between a life-form proposition and a general proposition appears to explain other forms of attribution different from defect. While perhaps still being accidents, certain kinds of accident seem more suited to being called attributions of some sort. For example, Thompson appears to propose an inference from a mismatch between life-form proposition and general proposition that explains attributing oddness to a member of a species. Thompson writes “Consider that we might explain a conception of, say, oddness, with some such rule as follows: From: “Most A’s are F”, and: ”This A is not F”, to infer “This A is odd in that it is not F”.” If we think life-form propositions of the form cA“Most A’s are F” are possible, then further simple attributive inferences seem reasonable to make as well. For instance, the following rule might be proposed to explain the conception of tallness. From:cS “S’s are x to y units of measurement high” and: “ N, a member of S, is y+ units high” conclude “N is a tall S.” While these rules probably need refinement as accounts, they show prima facie that other uses of Aristotelian categoricals exist.
We might be cautious in thinking Thompson really means to propose this as an account however, since his ‘oddness schema’ appears to have a strange consequence when considered with his ‘evaluative schema’. The schema for oddness as given by Thompson always commits us to connecting oddness with defectiveness. For if “Most A’s are F” and “This A is not F”, then it appears that the evaluative schema holds “This A is defective in that it is not F”. So, it appears that Thompson is committed to saying that all cases of oddness are also cases of defect. This result seems to clash with the intuition that a creature might be odd without being defective; nothing in the concept of oddness suggests that it is necessarily connected to a concept of defect. So in the interest of reading Thompson charitably, it is unclear whether the ‘oddness schema’ should be thought of as a serious proposal.
If Thompson’s oddness proposal is taken seriously, as I would like to take it, then we need to find a way of understanding the evaluative schema such that it is not committed to taking all odd things as defective things. I would like to point out here that my account of the discrepancy between life-form propositions and more particular propositions is not committed to connecting all accounts of oddness with defect, but rather, with some sort of accident. Taken with the accident schema, the oddness schema holds that if an organism is odd, then the organism differs from how it is supposed to be. This connection between concepts seems much more in line with our ordinary thoughts about oddness.
Finally, since Thompson seems willing to consider that some life-form propositions are of the form “Most A’s are F”, it seems more likely that he will accept the representation of life-form propositions as cspecies "x (Species(x) ’ F(x)), with first order quantified w.f.f.’s being operated on by the species-relative ‘supposed to’ operator. And since we know how to represent “Most” using quantifiers, adapting such a rule does not appear to present any particular difficulty.
Folk Biology
Given what has been said, one misinterpretation of Thomson’s theory so far is to think that the theory of Aristotelian Categoricals is some sort of folk biology. James Lenman, in his article “The saucer of mud, the Kudzu vine and the uxorious cheetah: against neo-Aristotelian naturalism in meta-ethics” appears to take such a position. Lenman writes:
“But folk biology is only a credible foundation for ethics or anything
else insofar as it is consistent with and can be subsumed within the
findings of the sort of biological understandings we have a sound
scientific warrant to believe are true. Insofar as folk biological thought
is pervasively and irreducibly teleological in the ways Foot, Thompson
and others emphasize, then our best science very plausibly tells us that
folk biology simply gets it wrong. “
James Lenman, “The saucer of mud, the Kudzu vine and the uxorious
cheetah: against neo-Aristotelian naturalism in meta-ethics”, pg 47
If Thompson’s arguments regarding Aristotelian categoricals are correct, then even our best science already makes use of the kind of thoughts Lenman labels as “folk biology”. For the biologist makes use of thoughts about life-form when she studies members of a life-form in the field, thoughts that affect her judgment about members. Whether some structure counts as an abnormality or a part of an organism does not depend on observing some number of creatures having the structure, but on the life-form of the creature. Similarly, Thompson has argued that the biologist would be wrong to think of some carbon structure as truly a leg, or a wing unless that carbon structure truly falls under a life-form.
Further, insofar as the biologist excludes data from defective or sick specimens in her study of the normal or healthy behavior of some species, she makes judgments as to which specimens are healthy and which are sick. As Thompson argues, such judgments are not based on statistical observations, for surely we will concede that it is possible for a statistical majority of a species to be sick or defective. If we think that the biologist’s judgment of a specimen’s health and defect might nonetheless be correct despite not being supported by statistical data, then we have reason to examine how that could be. The apparatus of life-form propositions could be thought as explaining more formally how a biologist reaches true conclusions about defect, and looking at such a model makes clear the kinds of principles needed to be held for such a conclusion to be true.
Finally, neither Thompson nor Foot claim for there to be teleological relations of the sort Darwin’s theory rules out. Thompson even goes so far as to write “The description of this sort of order has nothing to do with natural selection either; these propositions are in no sense hypotheses about the past. The elements registered in natural historical judgments and the interconnections registered in the natural teleological judgments all alike belong to that peculiar ‘present’ which we saw contains both ‘spring’ and ‘fall’ in winter, and ‘the seventh year of the cicada’s life-cycle’ even during the second.” So Lenman’s charges here against Thompson and Foot’s position seem based on a misunderstanding.
