Roderick Milton Chisholm is widely regarded as one of the mostcreative, productive, and influential American philosophers of the20^th Century. Chisholm worked in epistemology, metaphysics,ethics, philosophy of language, philosophy of mind, and other areas.His work constitutes a grand philosophical system somewhat in themanner of Leibniz or Descartes. Chisholm continually refined —and sometimes utterly revised — his views. He was a prolificwriter. The bibliography of his written work in [LLP] containscitations of 320 items, including journal articles, reviews, andbooks. His work in epistemology alone would probably guarantee hisposition as an outstanding figure in American philosophy. Yet he mademajor contributions in several areas of metaphysics and ethics aswell. As a result, it would be impossible to give a comprehensiveaccount of Chisholm’s system in a brief article. Thus, in thisarticle we attempt to present no more than a sketch of some ofChisholm’s most distinctive and important views. We do not claimto present his “final, authoritative view” in any area.Nor do we discuss the critical secondary literature concerning hisviews.

• 1. Biographical
• 2. Philosophical Method
• 3. Epistemology I — Epistemic Terms, Principles, Foundationalism
• 4. Epistemology II — Other Epistemological Doctrines
• 5. Metaphysics I — The Nature of Persons
• 6. Metaphysics II — Agency and the Free Will Problem
• 7. Metaphysics III — The Ontology of Events, States of Affairs, and Facts
• 8. Metaphysics IV — Brentano’s Thesis
• 9. Metaphysics V — “Chisholm’s Paradox”
• 10. Ethics I — Intrinsic Value, Defeat, Theodicy
• 11. Ethics II — Contrary to Duty Imperatives
• 12. Ethics III — The Ethics of Requirement
• 13. Evolution and Criticism of Chisholm’s Views
• Bibliography
  • Primary Literature
  • Secondary Literature
• Academic Tools
• Other Internet Resources
• Related Entries

1. Biographical

Chisholm was born in North Attleboro, Massachusetts in 1916. As anundergraduate, he studied philosophy at Brown University where heworked with a number of distinguished philosophers including C. J.Ducasse and R. M. Blake. He received his Ph.D. from Harvard Universityin 1942. At Harvard, Chisholm worked mainly with C. I. Lewis andDonald C. Williams. Bertrand Russell and G. E. Moore visited Harvardduring Chisholm’s time there and each of them evidently playedan important role in Chisholm’s development as aphilosopher.

After serving in the Army (primarily as a psychological tester) andgetting married, Chisholm was employed briefly as a lecturer at theBarnes Foundation in Pennsylvania. An amusing account ofChisholm’s experience at the Barnes Foundation can be found inChisholm’s intellectual autobiography in [LLP]. He then returnedto Brown as an assistant professor. He remained at Brown for the restof his long career (aside from periods as visiting professor atHarvard, Graz, Princeton, Chicago, Massachusetts, Salzburg, andseveral other places). He was Editor ofPhilosophy andPhenomenological Research from 1980 until 1986. He was thenAssociate Editor until the time of his death.

As a result of his dealings with Russell and Moore, Chisholm becameaware of the work of Brentano and Meinong. His interest in thesephilosophers eventually led to correspondence with Austrianphilosophers and visits to Graz. Several of Chisholm’spublications, perhaps especially “Sentences aboutBelieving” and [B&IV], helped to make some ofBrentano’s leading ideas familiar to American philosophers. In1972 Chisholm was awarded an honorary doctorate at Graz. He wasintrigued to discover that it would now be correct to address him as“Professor Doktor Doktor h. c. Roderick M. Chisholm.”

Chisholm was a tremendously successful teacher of philosophy. Hiscourses, both at the undergraduate level and at the graduate, werealways packed with enthusiastic students and colleagues. In spite ofhis great distinction, he was modest and amusing in the classroom. Heenjoyed engaging in animated critical discussion with students, andencouraged his students to present their questions and objections. Inmany cases these questions led to revisions of the doctrines Chisholmhad presented. He was always pleased to receive good criticism, andshowed enormous creativity in producing revisions in his attempts toovercome the problems. In his autobiography, Chisholm mentions howdelighted he was to have had such eager “refuters” in hisclasses at Brown as well as at other places where he visited. For atouching discussion of Chisholm’s teaching style, see Taylor(1975) .

Chisholm directed about 59 doctoral dissertations, thus making himperhaps the third most prolific producer of philosophy PhDs inAmerican history. Many of his students went on to have distinguishedcareers of their own. If we consider the class containingChisholm’s students and the students of those students, itbecomes obvious that through his teaching, Chisholm influenced aremarkable number of philosophers. In addition, many ofChisholm’s colleagues and friends turned their attention toChisholm’s work.

Chisholm published an extraordinary number of journal articles andreviews. A short discussion of some of the most important of these isincluded below in Section 13. He also edited, co-edited, andtranslated several works of others. Among the most important of thebooks written by Chisholm are:Perceiving: A PhilosophicalStudy [PPS],Theory of Knowledge ([TK1], [TK2], and[TK3] for first, second, and third editions),Person and Object: AMetaphysical Study [P&O],The First Person: An Essay onReference and Intentionality [FP],The Foundations ofKnowing [FK],Brentano and Intrinsic Value [B&IV],On Metaphysics [OM], andA Realistic Theory ofCategories: An Essay on Ontology [RTC].

Chisholm received a remarkable array of academic honors and awards. Hewas a Fellow of the American Academy of Arts and Sciences, Presidentof the Eastern Division of the American Philosophical Association,President of the Metaphysical Society of America, and a member of manyother distinguished councils and boards. He presented importantlectures at Oxford, London, Stanford, and elsewhere. A volume of theLibrary of Living Philosophers is devoted to his work.

Among the books devoted to Chisholm areAnalysis andMetaphysics (Lehrer 1975) [A&M],Roderick M.Chisholm (Bogdan 1986) [RMCp],Essays on the Philosophy ofRoderick M. Chisholm (Sosa 1979) [EPRMC], andThe Philosophyof Roderick M. Chisholm (Hahn 1997) [LLP]. A special issue of thejournalMetaphilosophy devoted entirely to Chisholm’swork was published in October of 2003.

Chisholm died in Providence, Rhode Island, in January of 1999.Memorial Minutes on Chisholm written by Ernest Sosa appeared in theProceedings and Addresses of the American PhilosophicalAssociation in May of 1999. Chisholm’s intellectualautobiography can be found in “My PhilosophicalDevelopment” in [LLP]. An earlier version of his intellectualautobiography appears as the “Self-Profile” in [RMCp].Some readers may be amused by Fred Feldman’s Banquet Addressfrom the Chisholm Memorial Conference held at Brown in November, 2000and available online (see Other Internet Resources).

2. Philosophical Method

Chisholm wrote and taught in a distinctive style that inspired hisreaders and students. His characteristic methodology was to begin hisdiscussion of a philosophical issue by identifying a few key questionsand citing pre-analytic data that an adequate theory shouldaccommodate. In many cases his work began in the Aristotelian fashionwith a set of “aporia” or puzzles. He sought to develop atheory that would be adequate to the puzzles. He formulated histheories by first introducing a small number of primitive orunanalyzed terms and then constructing an often elaborate system ofdefinitions and principles all built on these primitives. The finalprinciples and definitions were intended to provide the basis forsolutions to the puzzles with which he began. The clarity and eleganceof the systems were remarkable, though in some cases critics worriedthat the primitive concepts were for one reason or another suspect.Chisholm encouraged readers and students to criticize his systems byproposing counterexamples and objections. They were eager to do so,and Chisholm took great joy in revising and improving upon his viewsin the light of their comments.

Chisholm was well known for his penchant for formulating definitionsand subsequently revising them in the light of counterexamples. Theauthors of thePhilosophical Lexicon (see Other InternetResources) took note of this and accordingly introduced a newtechnical term of their own:

chisholm, v. To make repeated small alterations in adefinition or example. “He started with definition (d.8) andkept chisholming away at it until he ended up with(d.8′′′′′′′′).”

3. Epistemology I — Epistemic Terms, Principles, Foundationalism

Chisholm is perhaps best known for his work in epistemology. Thoughthe details of formulation steadily changed, Chisholm’sfundamental position in epistemology remained constant. Chisholm tookit as a starting point for his epistemological theorizing that we dohave knowledge of the external world. In addition, Chisholm acceptedfurther anti-skeptical doctrines concerning knowledge by memory of thepast, and a priori knowledge of some necessary truths. He did notassume that he couldrefute skepticism. Rather, he understoodthe central project of epistemology to be the project of showing indetail how it is possible for us to have quite a lot of the knowledgethat, in our reflective moments, we take ourselves to have. Inadopting this stance, he seems to have been influenced by Moore aswell as some of his teachers at Brown and Harvard.

Perhaps the most distinctive element of Chisholm’s theory ofknowledge is a set of tightly linked epistemic principles. These serveto display the relation between directly evident foundationalknowledge and indirectly evident beliefs about the external world, thepast, and other matters about which we can have knowledge. In hisdoctoral dissertation, Chisholm made his earliest attempt to formulatea set of such principles. [PPS] contains a more sophisticated systemof principles. In the three editions of [TK] Chisholm presented evenmore carefully worked out sets of proposed epistemic principles.

When he stated these principles, Chisholm made use of several terms ofepistemic appraisal. He appreciated the importance of explainingprecisely what each of these terms means. He also appreciated theimportance of explaining precisely how they are related. In an effortto explain all this, Chisholm started with a single primitiveepistemic concept — this is the concept ofgreaterreasonability, which relates the holding of one propositionalattitude toward some proposition with the holding some otherpropositional attitude toward some proposition. In his discussion ofthis concept, he pointed out that holding an attitude (belief, denial,withholding) in one proposition could be more reasonable for a certainsubject than holding some attitude (belief, denial, withholding) inanother proposition. And then, making use of this fundamentalepistemic concept (as well as the concepts ofbelief,refraining andnegation), Chisholm defined the conceptsofcertainty, beingevident, beingbeyondreasonable doubt, beingacceptable, and so on.

In [TK1] (22), Chisholm proposed these definitions of some centralterms of epistemic appraisal:

p isreasonable forS att=_df believingp is more reasonable forS att than withholdingp.

p isacceptable forS att=_df withholdingp is not more reasonableforS att than believingp.

p isevident forS att=_df (i)p is reasonable forS att, and (ii) there is no propositionq such that itis more reasonable forS to believeq attthan it is forS to believep.

Chisholm maintained that there are logical relations among theseconcepts. Thus, for example, he said that if something is evident,then it is (at least) reasonable. A key necessary condition forknowing a proposition, according to Chisholm, was that the propositionbe evident. In some of his early works, Chisholm analyzed knowledge asevident true belief. In subsequent works he modified this in responseto the Gettier problem.

Chapter 3 of [TK1] provides a good illustration of Chisholm’sway of using these terms of epistemic appraisal in the statement ofhis epistemic principles. In that chapter, he presents a set of nineepistemic principles. The first of these concerns“self-presenting states”. These are states such as belief,hope, fear and other propositional attitudes, as well as phenomenalstates of being appeared to in various ways, as well as states ofintending or undertaking to do something. The principle is:

(A) If there is a “self-presenting state” such thatS is in that state, then it isevident toSthat he is in that state.

Principle (A) would play a central role in a proposed explanation ofthe possibility of a certain form of introspective knowledge. Forexample, it would figure in an explanation of how it is possible for aperson to know that it now seems to him that he is seeing a doorknob.By itself, (A) has no implications for knowledge of the externalworld.

Subsequent principles purport to explain further sorts of knowledge.Principle (B) says:

(B) IfS believes he perceives something to have a propertyF, then the proposition that he does perceive something tohaveF, as well as the proposition that something hasF isreasonable forS. ([TK1], 45)

(B) applies only in cases in whichS thinks he is seeing, orsmelling, or in some other way perceiving something to haveF; it does not imply that in such conditions there actuallyis something that hasF, or that it is evident toSthat there is something that hasF. It implies merely that itisreasonable forS to believe this. Sincereasonability is not a sufficiently strong epistemic condition forknowledge, this would not explain how knowledge of the external worldis possible. But the next principle may seem to explain this.

(C) If there is a sensible characteristicF such thatS believes that he perceives something to haveF,then it isevident toS that he is perceivingsomething to haveF, and that there is something that hasF. ([TK1], 47)

Principle (C) applies only in cases whereF is a sensiblecharacteristic. This would be a feature that is appropriate to one ofthe senses, or a common sensible. But even when so restricted, (C)seems to be a very strong principle, for it seems to imply that someof our beliefs about objects in the external world are evident —and this would be sufficient for knowledge according to the analysisof knowledge that Chisholm accepted at the time.

However, on the next page Chisholm expresses some serious reservationsabout (C). He mentions an analogy to statements of prima facieobligation. Such obligations can be overridden. These later remarkssuggest that Chisholm had intended (C) to be understood as a principleof merelyprima facie evidence. Apparently, then, the realprinciple is:

(C′) If there is a sensible characteristicF such thatS believes that he perceives something to haveF,and this belief does not occur as part of some overriding widersituation, then it isevident toS that he isperceiving something to haveF, and that there is somethingthat hasF.

Principle (C′) is considerably less bold than principle (C).This can be seen if we reflect on what would count as a counterexampleto (C). Consider a case in which a person knows perfectly well that hehas just taken some hallucinogenic drugs. Suppose he knows the typicaleffects of these drugs. Suppose he has good reason to believe thatthere are no unicorns, but now he believes that he is perceivingsomething to be a unicorn.

Principle (C) seems to imply that it is evident to him that there is aunicorn. Principle (C′) does not have this implication. The factthat the appearance occurs within this wider context defeats its primafacie epistemic status. Although (C′) is less bold than (C), itdoes help to explain how we can have knowledge of the sensiblecharacteristics of external objects.

Then come three principles about memory:

(D) IfS believes he remembers having perceived something tohaveF, then the proposition that he does remember havingperceived something to haveF, as well as the propositionthat he did perceive that something hadF, and theproposition that something wasF, isacceptable forS.

(E) IfF is a sensible characteristic, andSbelieves he remembers having perceived something to haveF,then the proposition that he does remember having perceived somethingto haveF, as well as the proposition that he did perceivethat something hadF, and the proposition that something wasF, isreasonable forS.

(F) If there is a self presenting stateP such thatS believes he remembers having been inP, then theproposition that he remembers that he was inP, as well asthe proposition that he was inP, is one that isreasonable forS.

None of these principles would explain how we can have knowledge ofthe past, for none of these principles implies that beliefs about thepast can rise to the status of beingevident — andaccording to Chisholm a person can know a fact only if it is evidentfor him.

Chisholm wanted to say that we can have knowledge of things beyondthose we “see with our own eyes”. These would includethings that are confirmed by the evidence of our senses. As a steptoward achieving this, he has a principle about such things. Theprinciple makes use of the concept of theempiricallyacceptable. To find what’s empirically acceptable forS, just consider all those propositions that get a rating ofacceptable (or higher) by the aforementioned principles. Those areempirically acceptable forS att. Chisholm does notdefine confirmation, but assumes we understand it.

(G) Ifh is confirmed by the set of all things empiricallyacceptable forS att, thenh isacceptable forS att. ([TK1], 53)

But note that this raises such things up to the level of acceptabilityonly. This is still not enough for knowledge.

Consider a set of propositions that are logically consistent andindependent. No member is entailed by any combination of the others;no member is such that its negation is entailed by any combination ofthe others. Suppose each member is confirmed by the combination of allthe others. Then that set is “concurrent”.

(H) Ifh is a member of a concurrent set of acceptablepropositions forS att, thenh isreasonable forS att.

Principle (H) serves to raise the epistemic status of acceptablepropositions. When they are members of concurrent sets, suchpropositions become reasonable. This is a coherentist element inChisholm’s foundationalism in [TK1]. But it is still not quitesufficient for knowledge of the external world. For that, Chisholmfinally introduces:

(I) IfS believes at t that he perceives something to beF, andh is the proposition thatthere issomething having the property F, andh is a member of aset of concurrent propositions each of which is acceptable forS, thenh isevident toS att.

So now we can see how, according to Chisholm at the time of [TK1],knowledge of the external world is possible. Let’s take theproposition that there is a doorknob here. How can I know that?Suppose I am looking at, smelling, feeling, and remembering thedoorknob. Then via (B) lots of such propositions are reasonable forme. Suppose these all confirm the proposition,d, that thereis a doorknob here. Then via (G),d is acceptable. Supposeall of these together withd form a concurrent set. Then via(H), d is reasonable. Suppose I think I am seeing the doorknob, andthe proposition that there is a doorknob is a member of a concurrentset of acceptable propositions. Then it is evident to me that there isa doorknob here. If it’s true, then I know it. If all theseconditions are satisfied, I have knowledge of the external world.

Critics may complain that this set of principles is incomplete andsketchy. (Indeed, in spite of the principles about memory, nothing inthis system even purports to explain how someone could have knowledgeof the past.) Chisholm has given us no more than a sketch of apossible answer to the question of how empirical knowledge ispossible. But it reveals the foundational, coherentist, Chisholmianepistemic principle-ist structure of his thought. All later variantsretain these features. What is most distinctive about these principlesis that they are not instances of more general logical principles andChisholm does not claim that they are true in virtue of any factsabout causal connections or reliability. They are presented asfundamental epistemological facts.

4. Epistemology II — Other Epistemological Doctrines

According to Chisholm, epistemology consists of Socratic inquiry intothe questions “What can we know?” and “What are thecriteria of knowledge?” He thought that a puzzle faces anyonewho attempts to answer these questions. It appears that to answer thefirst question, one needs a criterion to distinguish between thingsthat are known and things that are not known. That is, one needs ananswer to the second question. But to have an answer to the secondquestion, he thought, one needs a list of the things one knows so thatone can identify the features that distinguish knowledge from itsopposite. That is, one needs an answer to the first question. Lackingsuch an answer, Chisholm feared, one would not be in a position to beconfident that any proposed criterion of knowledge was correct.Chisholm calls those who think that they have an answer to the secondquestion that they can use to answer the first“methodists” and those who think that they have an answerto the first question that they can use to answer the second“particularists.” Chisholm himself was a particularist,yet he claimed that he had no argument to offer against methodism oragainst the view that neither question could be answered without aprior answer to the other. In a number of places he said that theproblem of the criterion could be answered only by begging thequestion.

A final epistemological doctrine for which Chisholm is particularlywell-known is internalism. Chisholm characterized internalism in thefollowing way: ‘The internalist assumes that, merely byreflecting upon his own conscious state, he can formulate a set ofepistemic principles that will enable him to find out, with respect toany possible belief he has, whether he is justified in having thatbelief. The epistemic principles that he formulates are principlesthat one may come upon and apply merely by sitting in one’sarmchair, so to speak, and without calling for any outside assistance.In a word, one need only consider one’s own state of mind’[TK3: 76]. A crucial implication of this doctrine is that people whoseconscious states are alike must be justified in believing the samepropositions. Also present in the quoted passage is a related themeconcerning the autonomy of epistemology. Chisholm held thatepistemologists did not need the assistance of the empirical sciencesin answering their purely epistemological questions. In advancingthese doctrines, Chisholm took issue with the externalist andnaturalistic theories, such as the causal theory and reliabilism, thatgained favor with many epistemologists toward the end ofChisholm’s career.

5. Metaphysics I — The Nature of Persons

Throughout his career, Chisholm thought and wrote about themetaphysics of persons. His book,Person and Object, containsan extended discussion of the topic and the claims he makes thereprovide a good illustration of his views about persons. Chisholmmaintained that a fully satisfactory metaphysics of persons would haveto accommodate certain bits of preanalytic data, or else explain howthey can be false. He took these bits of data to be “innocentuntil proven guilty”. Among these is the notion that persons aregenuine objects, and not merely logical fictions. Talk about personsmust be taken seriously, and not as a mere figure of speech. Chisholmmentioned further data (which he presented in the first person): (1) Ihave various beliefs, feelings, desires, attitudes; (2) I have a body;(3) I am intentionally bringing about various things that I could haverefrained from bringing about. He went on to affirm past-tenseversions of (1)–(3): at various times in the past I had certainother beliefs, etc. Chisholm emphasized a kind of “unity ofconsciousness” thesis according to which the contemporaneouspsychological facts pertain to one and the same thing. Additionally,he thought he was entitled (at least until strong evidence to thecontrary is provided) to assume that there is one thing that hasbeliefs, and also has a body, and also engages in intentionalaction.

Chisholm argued that many of the alternative conceptions of personsdid not conform to the preanalytic data. One prominent alternative isthe so-called “Bundle Theory”, which illustrates astrikingly different conception of persons. On this view, there is nothinker of my thoughts, no “owner” of my body, and no doerof my deeds. Instead, there is just a “bundle ofperceptions”. Thelocus classicus of this view is awell-known passage in Hume’sTreatise where hesays:

For my part, when I enter most intimately into what I call myself, Ialways stumble on some particular perception or other, of heat orcold, light or shade, love or hatred, pain or pleasure. I never cancatch myself at any time without a perception, and never can observeany thing but the perception. When my perceptions are removed for anytime, as by sound sleep; so long am I insensible of myself, and maytruly be said not to exist. (Treatise of Human NatureI.iv.vi)

Chisholm points out that this view is in sharp conflict with thepreanalytic data since, if taken seriously, it implies that there isnothing that is having the perceptions, or feeling the pains andpleasures. He goes on to claim that Hume’s argument for theBundle Theory is based on a bit of evidence that presupposes (contraryto the theory) that there is an object that is doing the perceiving,and thinking the thoughts. Chisholm points out that Hume mentionsoccasions on whichhe enters into what he calls himself; Humesays that on these occasionshe stumbles on some particularperception; he says thathe never notices anything other thanthe perceptions. Who or what is this thing that is doing all thisentering and stumbling and failing to notice? Is it some single thing?Does it persist through the time of entering and stumbling and failingto notice? If it does, then it is the very thing whose existence Humeseems to be concerned to reject.

Certain forms of materialism have been endorsed by many philosophers.According to some of these theories, each person is to be identifiedwith his or her body. The body is the bearer of both physical andmental properties. The person persists so long as his or her bodypersists. Chisholm, however, was always uneasy about such theories.One of his most distinctive lines of objection involves the use ofsome metaphysical distinctions. Consider a ship that persists throughsuccessive replacements of parts. First one plank is replaced, thenanother, until after some time many of the ship’s original partsare no longer present. Chisholm describes the situation by saying thatat each time that the ship exists, it exists in virtue of the factthat a certain collection of parts exists then. The collection ofparts that exists at a moment is an “ens per se”— it exists in its own right; but the ship is an “ensper alio” — it exists in virtue of the fact that someother things exist. Different collections of parts do duty for, orstand in for, the ship at different times. In some cases, a thing hasa property at a time in virtue of facts about happenings at othertimes. Thus, for example, a person might be a widow at a certain time,but that’s because of marriages and deaths that happenedearlier. Other properties are not in this way “rooted” infacts about other times. With respect to properties that are notrooted outside the times they are had, Chisholm makes an interestingclaim about a difference between things that exist in their own rightand things that exist in virtue of others. The ship (anens peralio) has properties at times in virtue of the fact that itsstand ins (entia per se) have those properties at thosetimes.

A famous old argument reveals something important about the identityof the ship through time. Suppose a ship (call it “Theseus1”) was constituted by a certain collection of parts at acertain time long ago. Suppose the parts in that collection weregradually replaced by newer parts. Eventually we find ourselves with aship (“Theseus 2”) that is the result of all this gradualreplacement. It is the apparent “descendant” of Theseus 1.Suppose, however, that the cast-off parts were all saved andeventually were reassembled so as to constitute a ship (call it“Theseus 3”). Suppose Theseus 3 contains all and only theparts that constituted the original ship Theseus 1 at the outset.Since Theseus 3 is part-for-part indiscernible from Theseus 1, thereis an inclination to identify Theseus 3 with Theseus 1. But sinceTheseus 2 emerged from Theseus 1 by a series of individually minorreplacements, there is also an inclination to identify Theseus 2 withTheseus 1. We cannot consistently give in to both of theseinclinations since it is obvious that Theseus 2 is distinct fromTheseus 3. Chisholm suggests that the facts in the world leave thissomewhat indeterminate. If a case like this were to arise in the realworld, we would leave it to the courts to make a determination ofwhich ship is the legitimate successor to the original. If the courtsdecide that Theseus 2 is the “real” descendant of Theseus1, then we will have to accept this conclusion for all practicalpurposes.

When a thing — like the persisting Ship of Theseus — is alogical construction out of more fundamental entities, questions ofidentity through time become somewhat conventional. Thus Theseus 2 canbe identical with Theseus 1 only in a “loose and popular”sense of ‘identical’. In the “strict andphilosophical” sense a collection of parts, C1, can be identicalwith a collection of parts, C2, only if C1 and C2 contain exactly thesame parts. In this strict sense, Theseus 2 is not identical withTheseus 1.

At least some of what has been said about the Ship of Theseus holdsalso for human bodies. Old parts are steadily being replaced by newones. After a period of time the body may contain little of the stuffthat originally constituted it. As a result, the body of an old personmay be said to be identical to the childish body with which he startedonly in the loose and popular sense. This has implications formaterialism.

If some typical form of materialism is true, then each person is hisor her body. The implication is that an old person may be said to beidentical to himself as a child only in a loose and popular sense. Ifwe take this sort of materialism seriously, we will have to say thatno person strictly persists through any change of parts. Chisholmfinds this incredible. He recalls an argument attributed to C. S.Peirce: suppose you are about to undergo surgery. Suppose you are toldthat you can save a lot of money by taking the surgery withoutanesthesia. Instead, you will be given amnesia-inducing drugs thatwill make you forget the pain afterwards. Still you might be uneasy,since you might fear that the man undergoing the painful surgery willbe you. Chisholm then continues:

Suppose that others come to you — friends, relatives, judges,clergymen — and they offer the following advice and assurance.“Have no fear,” they will say. “Take the cheaperoperation and we will take care of everything. We will lay down theconvention that the man on the table is not you, Jones, but isSmith.” Whatought to be obvious to you, it seems tome, is that the laying down of this convention should have no effectat all upon your decision. For you may still ask, “Butwon’t that person be I?” and, it seems to me, the questionhas an answer. ([P&O]: 111)

Chisholm’s Peircean argument has considerable intuitive force.It surely seems to many of us that our persistence through time is notsimply a matter of convention. No matter what the courts decide,either the person undergoing the painful surgery will be you, or hewon’t be you. If this intuition is right, then materialism ofthe imagined kind is false. For materialism implies that you are anens per alio, a logical construction. Yet each of us takeshimself or herself to be anens per se.

In several places (e.g., 1991, 168), Chisholm hinted at a modalargument against materialism. He noted that he himself could survivethe loss of his hand. Thus, he has this modal property: being suchthat it is possible for him to exist without the hand that he actuallyhas. But Chisholm contended that his body was identical with theparticular collection of human bodily parts that then constituted hisbody. That collection of parts could not survive the loss ofChisholm’s hand. If Chisholm’s hand were annihilated, thatcollection would cease to exist. Thus the collection of parts does nothave the property of being such that it is possible for it to existwithout the hand that Chisholm actually had. One application ofLeibniz’s Law implies that Chisholm is not the collection ofparts that then constituted his body. The argument clearly presupposesthe doctrine of mereological essentialism, according to which if acompound thingW has a certain partP, thenW cannot exist without havingP as a part.

In yet other places ([P&O], 104), Chisholm pursues another line ofthought. Note for example that if a certain ship (anens peralio) has a property such as the property of weighing so-and-somany pounds at a certain time, it has this property in virtue of thefact that theens per se that constitutes it at that timeactually has that property. The “stand-in” really has theproperty; the logical construction “borrows” the propertyfrom the stand-in. But Chisholm considers himself at a time when hehas the property of hoping for rain. He asks whether he himselfactually has that property, or whether instead he just borrows it fromsome stand-in. He dismisses this suggestion, saying that it ‘isnot to be taken seriously’. If there are two things present (thepersistent Chisholm and the temporary stand in), and each of them ishoping for rain, then surely it is Chisholm himself who morefundamentally has the property and his stand-in who has itderivatively. In this case, Chisholm himself must be anens perse.

Perhaps persons are somehow simple entities. It might appear that theycould be simple bodily substances — perhaps like extra-smallversions of the Luz Bone mentioned by Leibniz (in Book II, Chapterxxvii of theNew Essays). But Chisholm had defined‘bodily substance’ as ‘substance with parts’.Thus on his view, there cannot be a simple bodily substances. So ifthe person is a simple substance, it must be simple and immaterial— a “monad” as Chisholm says.^[1]

In “On the Simplicity of the Soul” (1991), Chisholmpursues this Leibnizian line. He identifies a sort of property that hecharacterizes as “qualitative”. Any such property wouldcharacterize substances (rather than other properties, or states, orabstract objects); it would be a property that could characterize asimple thing; it would be an internal (or, as some might prefer tosay, “intrinsic”) property of anything it characterized– thus a thing existing in isolation could have such aqualitative property. Chisholm says that the familiar mentalproperties that we know ourselves to have (e.g., the property ofhoping for rain) are in this way qualitative. So we know that we havequalitative properties. But, as he says, we never know, with respectto any compound thing, that it has any such qualitative property.Thus, there is reason to doubt that we are compound things. We knowourselves to have properties of a kind such that we do not know, ofany compound thing, that it has properties of that kind. It’snot entirely clear that this argument establishes that people are notcompound things. Perhaps it shows only thatfor all we knowwe might be simple entities.

6. Metaphysics II — Agency and the Free Will Problem

We like to think that we are sometimes morally responsible for some ofthe things we have done. It seems natural to suppose that if a personis morally responsible for something he has done, then he could havedone otherwise. But there is a puzzle here. Precisely what do we meanwhen we say of a person who acted in a certain way that “hecould have done otherwise”? Chisholm takes this as a fundamentalquestion in the metaphysics of action. It seems to be his way ofconceiving of the so-called “free will problem”. Hediscussed this topic in “Freedom and Action” (1966),“He Could Have Done Otherwise” (1967a) and then again in[P&O] as well as in quite a few other papers. It would be verydifficult to describe in detail all the twists and turns ofChisholm’s constantly evolving views on this topic. Instead, wewill focus on some of the more persistent and characteristicallyChisholmian points.

Suppose a certain person in fact did not go to Boston this morning,but suppose it seems that he could have done so. We want to say thatthough he stayed in Providence, he could have done otherwise —he could have gone to Boston. Surely the statement that he could havedone otherwise does not mean merely that it is logically possible thathe did otherwise, for superhuman and miraculous bits of behavior arelogically possible. But we don’t want to say, of a person whofailed to perform a miracle, that he could have done otherwise. Nordoes our statement mean that it is epistemically possible that theperson did otherwise. Suppose that this man was confined inProvidence, and utterly incapable of getting to Boston on a certainmorning. Suppose we know nothing about this confinement. We just knowthat he remained in Providence. Under these circumstances, while it isepistemically possible for us that he went to Boston, it would not becorrect to say that he could have done so. His doing otherwise thanstaying in Providence was epistemically possible for us, but notpossible for him in the way that bears on moral responsibility.

Some have said that ‘can’s are constitutionally iffy. On asimple version of this view, to say that someone could have doneotherwise is just to say that if he had chosen to do otherwise, thenhe would have done otherwise. Chisholm points out ([P&O],56–7) two reasons why this is wrong. First, suppose the personwas capable of traveling in any direction and easily could have goneto Boston; but suppose in addition that he did not know the way toBoston. If he had chosen to go to Boston, he would have ended up inNew London. Then it is correct to say that he could have gone toBoston, but incorrect to say that if he had chosen to go to Boston, hewould have done so. Secondly, suppose the person is incapable ofchoosing to go to Boston. Maybe he is overwhelmed with fear of Boston.But if nothing else prevents the trip, it will be correct to say thatif he had chosen to go to Boston, he would have gone, but it isincorrect to say that he could have gone to Boston.

Another account of ‘could have done otherwise’ makes useof the concept ofsufficient antecedent causal condition. Wemight think that when we say that someone could have gone to Bostoninstead of staying in Providence, what we mean is merely that at someearlier time this morning, his trip to Boston was causallyindeterminate — there was no sufficient causal condition eitherfor his going to Boston or for his not going to Boston. Chisholmargues against this idea, too. Suppose another person was lying inwait in Chelmsford. Suppose this other person would have freelyinterfered with our man’s travel plans if he had tried to get toBoston. Then it would not be correct to say that the man could havegone to Boston, but it would have been correct to say that there wasno sufficient causal condition then in place that would have preventedthe trip. The relevant sufficient causal condition would not havearisen until the person lying in wait in Chelmsford had freelyinterfered.

Thus we have a problem. What do we mean when we say that a personcould have done otherwise? Chisholm begins by introducing someconcepts that will serve as conceptual primitives in his account. Thefirst of these is a concept that Chisholm calls “causalcontribution”. If one event helps to bring about another, thenwe have one familiar sort of causal contribution. But in other casesit is not an event, but a person, who causally contributes to someevent (1966, 284). A case (if there are any) in which a personcausally contributes to some event would be described as an instanceof “agent causation”. Chisholm claims that in an earlierera philosophers would have thought that agent causation is thefamiliar notion; they would have assumed that event causation is to beexplained in terms of it. He recognized that at the time of hiswriting the priority relation between the concepts had probablyreversed.

The second concept that Chisholm employs is the concept ofundertaking orendeavoring. This is an irreduciblyteleological concept. It introduces the notion of purpose, or aim.Chisholm makes use of this undefined expression: ‘SmakesB happen in the endeavor to makeAhappen’. Suppose, for example, that a man dials a certaintelephone number intending thereby to make a certain phone ring in LosAngeles. (Recall that Chisholm started working on this material backin the 1960s. Telephones had dials in those days.) Let Smith be theman; letB be the state of affairs that consists inSmith’s dialing the phone; letA be the state ofaffairs that consists in the phone’s ringing in Los Angeles.Then it would be correct to say, using Chisholm’s undefinedtechnical term, that Smith madeB happen in the endeavor tomakeA happen.

From the fact that Smith thus endeavored to make the phone ring, itdoes not follow that phone actually rang. Maybe it was off the hook.Thus endeavor has a certain intentional aspect. Furthermore, assumethat the telephone in Los Angeles is in fact the only purple phone inthe Area Code. Even if Smith dialed in the endeavor to make that phonering, it does not follow that he dialed in the endeavor to make theonly purple phone in the Area Code ring. Maybe Smith didn’t knowthe color of the phone; maybe he didn’t care about making purplephones ring. So the concept of endeavor has a further intentionalfeature.

Chisholm then introduces an element of indeterminism: suppose that ata certain time there is no sufficient causal condition for a person toendeavor to bring aboutp; suppose in addition that there isthen is no sufficient causal condition for him to fail to endeavor tobring aboutp. Then Chisholm wants to say that he is free toendeavor to makep happen.

In some such cases a person might be free to endeavor to bring about acertain state of affairs, and furthermore if he were to endeavor tobring it about, he would succeed. Such things are directly in aperson’s power. Chisholm notes that he is thus making theconcept of power “constitutionally iffy”.

In other cases a person might not have a certain state of affairsdirectly in his power, but there might be a sequence of things,<p,q,r,s> such that hehasp directly in his power, and if he were to bring aboutp, he would then haveq directly in his power, andif … he would haves directly in his power. In such acase, the person hass indirectly in his power.

With these concepts in place, Chisholm is prepared to answer thequestion with which he started. Suppose again that a certain persondid not travel to Boston this morning. To say that he could have doneotherwise — that he could have traveled to Boston this morning— is just to say that as of this morning, going to Boston waseither directly or indirectly in his power. Chisholm claims —with some plausibility — that this account of the meaning of‘he could have done otherwise’ explicates what those whobelieve in freedom have in mind when they say that someone could havedone otherwise. Furthermore, in virtue of various technicalities ofthe definitions, it apparently turns out that Chisholm’s accountof the meaning of ‘he could have done otherwise’ isadequate to all the cases that prove difficult for theories such asthe ones surveyed above.

One essential element of Chisholm’s view is the idea that thereis a distinction between event causation and agent causation. In casesexclusively involving event causation, everything that causallycontributes to an event is another event. In cases involving agentcausation, among the things that causally contribute to an event is aperson. In virtue of this feature, Chisholm’s view is generallycategorized as a form of libertarianism. Another essential element ofhis view is the idea that when a person acts freely, he does somethingfor which there is no event or state or combination of events orstates that is a sufficient causal condition for his doing it. (Thereis also no event or state that is a sufficient causal condition forhis failing to do it, obviously.) In virtue of this feature of hisview, it may be seen as incorporating an element of indeterminism.

7. Metaphysics III — The Ontology of Events, States of Affairs, and Facts

Consider the following list of kinds of things:

1. Concrete, individual objects, or “substances”
2. Properties, attributes, qualities, relations
3. Times
4. Places
5. Propositions
6. Concrete, individual events, or “happenings”
7. Facts
8. Sets

We can imagine a metaphysician making a grand claim about this list.He could say that there really are things in each of those eightcategories. Furthermore, he could say, there is no redundancy in thelist: there is nothing that belongs in more than one of thecategories. Finally, this metaphysician could claim that the list iscomplete: everything that exists falls into one or another of thecategories in the list. In this way our metaphysician would haveendorsed a fairly robust ontological scheme.^[2]

Throughout quite a long period of time, Chisholm would have maintainedthat the imagined ontological scheme violates Occam’s Razor,since it affirms the existence of more fundamental kinds of thingsthan we really need. More specifically, he would have thought that hecould accomplish considerable pruning by replacing three of theseproposed categories with a single new category — “statesof affairs”. He thought he could explain all talk ofpropositions (category 5) by appeal to states of affairs of a certainsort. Similarly, he thought he could explain all talk of concreteevents (category 6) by appeal to states of affairs of a differentsort. Facts (category 7) would turn out to be nothing more than truepropositions.

Underlying Chisholm’s efforts here was a methodological viewabout ontological commitment. Suppose there is a certain statementthat we take to be true; suppose that statement seems to require theexistence of entities of a certain type. We then must admit entitiesof that type into our ontology unless we can find a way ofparaphrasing the statement in such a way as to preserve our originalmeaning, but to avoid any implication of the existence of the entitiesin question. As Chisholm puts it:

… from the fact that a true sentenceseems to commitus to a certain type of thing it does not follow that there is in factthat type of thing. For perhaps what the sentence tells us can bere-expressed in such a way that it no longer even seems to commit usto the type of thing in question. ([P&O], 116)

Thus, Chisholm undertook to provide paraphrases for sentences aboutconcrete events, propositions, and facts. In each case, he provided aparaphrase that involved states of affairs (and some other items, suchas times). But since the project turns crucially on various claimsabout states of affairs, Chisholm took special pains to explainprecisely what he took a state of affairs to be.

As examples of states of affairs, Chisholm mentions such things asthe author of Marmion being knighted andthe author ofWaverly being knighted. Chisholm takes these to be two differentstates of affairs; each of them existing necessarily; neither of themdepending for its existence on the existence of any contingent thing.Thus, each of these states of affairs would exist even at a possibleworld in which there is no author of Marmion or Waverly. Furthermore,according to Chisholm, states of affairs mayoccur,obtain, ortake place. They serve as the objects ofsuch attitudes as belief, hope, and mere consideration. Some of themserve as causes and effects. In affirming the existence of states ofaffairs, Chisholm took himself to be aligned with Frege (who,according to Chisholm, used the term ‘Gedanke’for entities of this sort).

Chisholm did not want to define states of affairs as things that canpossibly occur, since he thought that some of them could not possiblyoccur. So he proposed instead to say thatp is a state ofaffairs if and only if it is possible that there is someone whoacceptsp. Even if it’s impossible forp tooccur, it is still possible for there to be someone who accepts it. AsJaegwon Kim pointed out (Kim 1979), there is something a bit odd aboutthe fact that Chisholm attempted to define ‘state ofaffairs’ at all, since this is supposed to be a fundamental itemin his ontology. Furthermore, as Kim also pointed out, the definitionis of questionable value, since there is considerable disagreementabout the nature of the items that we “accept”. Some takethem to be sentences; others may take them to be mental items of some sort.^[3] Chisholmian states of affairs are presumably intended to be neitherlinguistic nor psychological.

For Chisholm,entailment is not a merely logical notion. Ifp entailsq, then not only doesp logicallyimplyq, it must also be the case, necessarily, that whoeveracceptsp also acceptsq. Chisholm makes use of thisdistinctive “intentional” concept of entailment to explainthe identity conditions of states of affairs. Wherep andq are states of affairs,p is the same state ofaffairs asq if and only ifp entailsq andq entailsp. This explains why Chisholm said thatthe state of affairs ofthe author of Marmion being knightedis distinct from the state of affairs ofthe author of Waverlybeing knighted, even if the author of Waverly happens to be theauthor of Marmion. So, for Chisholm, states of affairs are veryfine-grained entities.

Consider the state of affairs that is expressed by the sentence‘Someone is walking’. Chisholm wanted to say that thisstate of affairs occurs whenever someone walks, and fails to occur attimes when no one is walking. Other states of affairs are not likethis. For them, it is impossible to sometimes occur and sometimes failto occur. Chisholm claims that this provides the opportunity for anontological reduction. We can define aproposition as a stateof affairs of this latter sort — it is impossible for there tobe times when it occurs and other times when it does not occur. Atrue proposition is thus one that occurs; and afalse proposition is one that does not occur. Chisholm thinksthat we may understand the principles of logic to be about thesepropositions. By saying that afact is a true proposition,Chisholm gains yet another ontological reduction ([P&O], 123).

Chisholm thought that in some cases it makes sense to speak of thelocation at which a state of affairs occurs. Suppose John walks inChicago at a certain time. Then Chisholm would be willing to say thatthe state of affairs ofJohn’s walking occurs inChicago and at that time. Furthermore, some states of affairs entailcertain properties. The state of affairs ofJohn’swalking entails the property ofwalking because (i) ifthe state of affairs occurs, then something has the property ofwalking, and furthermore, whoever accepts the state ofaffairs believes that something has the property. This gives Chisholmthe ammunition he needs to explain what anevent is andthereby to effect another ontological reduction.

Among states of affairs, there are some that (i) occur, and (ii) arenot propositions. Furthermore, among these states of affairs, some(iii) entail contingent intrinsic properties that may be exemplifiedonly by individual things.^[4] Chisholm says that anevent is a state of affairs thatsatisfies all of these conditions. These are intended to play the rolein Chisholm’s ontology that concrete individual events play inthe ontologies of philosophers who accept concrete events asfundamental entities.

This way of conceiving events yields certain (possibly surprising)results. For one, every event occurs. For another, events happen attimes and places. Furthermore, as Chisholm sees it, when an eventoccurs, there must be some individual thing or things somehowparticipating in the occurrence of the event. These are the thingsthat have the properties entailed by the event. The event is said tobe “concretized” by those things. Finally, events mayrecur.John’s walking occurs when and whereJohn walks; it occurs only if John has the property ofwalking. It can occur on many different occasions and in manydifferent places. It may be an effect of some causes, and it may be acause of some effects. Chisholm goes on to discuss the role of eventsin perception, causation, and explanation.

He presents an ingenious system for counting events (“John wasinjured while taking his third walk of the day”) that makes useof the concepts of time, place, and concretization. Chisholm’sefforts in this project were described by one of his philosophicalopponents as “breathtaking” (Davidson 1970, 31).

We can see, then, that around the time of [P&O], Chisholm probablywould have accepted an ontological scheme more like this:

1. Concrete, individual objects, or “substances”
2. Properties, attributes, qualities, relations
3. Times
4. Places
5. States of affairs (some of which are event-like; some of which areproposition-like; and some of which are fact-like)
6. Sets

As time went by, Chisholm became dissatisfied with this picture. Onemain source of difficulty, as he saw it, was the commitment totimes. Furthermore, in spite of their utterly central placein this ontology, he began to feel uneasy about states of affairs.Some of his concern evidently arose as a result of conversations withhis colleague Jaegwon Kim, who had developed a different theory ofevents. On Kim’s theory, if we have a suitable objecta, propertyF, and timet, andahasF att, then there is an event,a’shaving F at t. An event is the having of aproperty by a thing at a time.

In his paper “Events without Times: An Essay in Ontology”(1990), Chisholm presented a new and even more sparse ontologicalscheme. As the title indicates, times were banished. Furthermore,states of affairs disappeared, replaced by the new category ofstates. A Chisholmian state is in some ways like a Kimianevent. If we have a suitable contingently existing object,a,and we have a suitable property,F, then, ifaactually hasF, then there is the statea’sbeing F. A state such as this is an event. Ifadoesn’t haveF, then there is no such state. There areno non-occurring states.

There are several things to note about these events:

First, Chisholm uses the terms ‘substrate’ and‘content’ to indicate the object and property involved inan event. The substrate must be a contingently existing thing. Thecontent must be a property that the substrate has contingently. Eachevent is said to have its substrate and content essentially. So ifthere was such an event as Plato’s writing of theRepublic, then that event essentially involved Plato and theproperty of writing theRepublic. The event itself could nothave occurred with a different substrate or content. Nor could it havefailed to have its “categorial” property ofbeing anevent.

Second, since Chisholm has banished times from his ontology, there areno times in a Chisholmian state. In this respect, Chisholm’sview differs from Kim’s. This might seem to leave open thepossibility that a Chisholmian state could occur for a while, thenstop occurring, and then start occurring again. If a state could dosuch a thing, then states would be relevantly like the now-banishedstates of affairs. But Chisholm denies this possibility. He says thatif he was reading earlier, then stopped reading, and then startedreading again, we havetwo distinct states with Chisholm assubstrate andreading as content. Since there are no“times” Chisholm is careful to avoid saying that whatdistinguishes these two states is their time of occurrence. When wesay, informally, that an event recurs, all we mean is that itssubstrate has had its content before.

Third, by the time of “Events without Times”, Chisholm hadbecome a presentist. He also was taking tense very seriously. To saythat there is such a thing asx is equivalent to saying thatthere is such a thing asxnow. On this view itwould be wrong to say that there is such a thing as Socrates; equally,it would be wrong to say that Socrates is such that he formerlyexisted but no longer does. Nevertheless, Chisholm constructedingenious (sometimes convoluted and unintuitive) paraphrases thatenable us to say everything we need to say about Socrates and othersno longer among us.

So the new ontological scheme looks more like this:

1. Concrete, individual objects, or “substances”
2. Properties, attributes, qualities, relations
3. States; some of which are event-like
4. Sets

[OM] contains a final summing up of Chisholm’s metaphysicalsystem. In [OM] Chisholm pares away still further. Among other things,he proposes to get rid of sets in favor of attributes.

8. Metaphysics IV — Brentano’s Thesis

Consider the statement that John is thinking about a unicorn. Thiscould be true even if no unicorn actually exists. But surely John isthinking aboutsomething. His mind is not simply a blank.Some philosophers have wanted to say in such a case that there is anobject — in fact, a unicorn – about which John isthinking, but that this object is one that has “intentionalinexistence”. It exists merely as an object of thought. InPsychology from an Empirical Standpoint, Brentano said thatevery psychical phenomenon is characterized by “intentionalinexistence”. He went on to say that this intentionalinexistence is “exclusively peculiar to psychical phenomena. Nophysical phenomenon shows anything similar”.

Brentano’s Thesis not entirely clear, but its importance shouldbe obvious. Suppose that every psychological fact somehow involvessomething with intentional inexistence (however precisely this is tobe construed). Suppose that no physical fact involves a thing withintentional inexistence. Then it would follow that no psychologicalfact can be identified with a physical fact; one version of thepsycho-physical identity theory would be false; physicalism of somepopular varieties would be untenable.

Chisholm was fascinated by Brentano’s Thesis. He published aseries of papers on the topic, starting in 1952 and carrying on untilat least 1981. In his papers Chisholm formulated precise versions ofBrentano’s thesis. Each paper provoked critical reaction fromother philosophers. Since in some cases the criticisms were cogent,Chisholm was motivated to develop new and more ingenious versions ofthe Thesis.

Tracing all the different ways in which Chisholm attempted toformulate Brentano’s Thesis would require a book-lengthdiscussion. Instead, we here focus on one illustrative example. InChapter 11 of [PPS], Chisholm developed a version of Brentano’sThesis that turned on the notion of anintentional sentence.He identified three marks of intentionality. Any noncompound sentencethat manifested one of these marks would be identified as anintentional sentence. The first mark of intentionality recallsBrentano’s idea of intentional inexistence. Consider thesentence ‘John is thinking about the Loch Ness Monster’.Neither this sentence nor its negation entails either that the Monsterexists or that it does not exist. Any such simple declarative sentenceis intentional.

The second mark of intentionality is closely related, but involveswhole propositional clauses rather than mere singular terms. Considerthe sentence ‘John wonders whether there is a monster in LochNess.’ Neither this sentence nor its negation entails eitherthat there is a monster in Loch Ness or that there is not such amonster. Any simple sentence that displays this feature is alsointentional.

The third mark of intentionality involves the failure ofsubstitutivity for coextensive singular terms. In fact Eisenhower wasthe man who would succeed Truman as President of the United States.However, from the fact that we knew that Eisenhower was in command ofthe Army, it does not follow that we knew that the man who wouldsucceed Truman was in command of the Army.

These three marks of intentionality apply only to noncompoundsentences. A compound sentence counts as intentional if and only ifone of its components is intentional. Thus, ‘if Parsifal soughtthe Holy Grail, then he was a Christian’ is intentional becauseits antecedent is intentional. The antecedent carries the first markof intentionality because neither it nor its negation entails eitherthat the Holy Grail exists or that it does not exist.

Chisholm then says — somewhat cautiously — that a thesis“resembling that of Brentano” may be stated in this way:we can express all of our beliefs about physical phenomena without theuse of intentional sentences; but when we describe psychologicalattitudes, then we must either use intentional sentences or else makeuse of other terms that are not needed for the description of physicalphenomena. In his discussion, it appears that when Chisholm speaks of‘terms that are not needed for the description of physicalphenomena’ he has in mind terms that are ultimately definableonly by the use of intentional sentences.

Chisholm recognizes that we sometimes use intentional sentences todescribe nonpsychological facts. For example, consider the sentence‘This weapon, suitably placed, is capable of making it be thecase that Boston is destroyed.’ Neither that sentence nor itsnegation entails either that Boston is destroyed or that Boston is notdestroyed. So it carries the second mark of intentionality. But itseems not to be psychological. Chisholm claims that we can rephrasethe sentence nonintentionally: ‘If this weapon were suitablyplaced, then Boston would be destroyed.’ This is a compoundsentence. Reflection will reveal that neither component carries any ofthe marks of intentionality. (The critic may wonder whether this replyis entirely satisfactory, since it may seem that the paraphrase doesnot mean precisely what the original meant.)

Another difficult class of sentences involves statements ofprobability. Consider this example cited by David Sanford: ‘Itis not probable that Providence will be hit by a comet in the21^st Century’. Neither this sentence nor its negationentails either that Providence will be hit or that it won’t behit; thus the sentence carries the second mark of intentionality. Yetit seems to be nonpsychological.

At the time of [PPS] Chisholm already recognized that this formulationof Brentano’s Thesis was problematic. He mentioned‘sentences describing relations of comparison’ as adifficulty. Consider the sentence ‘This lizard looks somethinglike the Loch Ness Monster’. Neither that sentence nor itsnegation entails either that the Loch Ness Monster exists or that itdoes not exist. Yet there seems to be nothing psychological about thestatement that the lizard looks something like the Monster.

Over a period of more than thirty years, Chisholm formulated andreformulated Brentano’s Thesis. Perhaps the most insightfuldiscussion of his views and their connections can be found in DavidSanford’s ‘Chisholm on Brentano’s Thesis’. Inreply to that essay, Chisholm said, ‘Anyone who wants tounderstand what I have been up to in trying to formulate criteria ofintentionality, should read Sanford’s paper’ ([LLP],215).

9. Metaphysics V — “Chisholm’s Paradox”

Chisholm’s paper “Identity through Possible Worlds: SomeQuestions” (1967b) was the lead article in the first issue thenew journalNoûs in 1967. In that article, Chisholmpresented an argument that provoked a remarkable amount of discussionand debate. A variant of the argument was subsequently presented inChisholm’s “Parts as Essential to Their Wholes”,which was Chisholm’s Presidential Address at the MetaphysicalSociety of America in 1973. This essay corresponds closely to AppendixB of [P&O]. The puzzles presented in those papers have come to beknown as versions of “Chisholm’s Paradox” (though itmust be pointed out that the term ‘Chisholm’sParadox’ is also widely taken to refer to the problem aboutcontrary to duty imperatives that is discussed below in Section11).

Chisholm’s target in the 1967b paper was a popular collection ofviews about the metaphysics of modality. Many philosophers —going back perhaps to Leibniz — seem to presuppose these viewsin their discussions of necessity and possibility. The centraldoctrine is that there are many possible worlds. The actual world isjust one among them, distinguished by the fact that what happens hereis what actually happens. To say that a statement istrue isjust to say that it is trueat the actual world. To say thata statement ispossible is just to say that it is trueatsome possible world. To say that a statement is necessary is justto say that it is trueat every possible world. Thus we havean account of de dicto modality.

On one typical construal, it is assumed that an individual who existshere at the actual world also exists at many other possible worlds.When we say that a certain individualactually has a certainproperty, we mean that he has that property at the actual world. Tosay that hepossibly has that property is just to say that hehas that property atsome possible world. To say that henecessarily has that property is just to say that he has thatpropertyat every possible world where he exists at all. Thisgives us an account of de re modality. But it also generatesquestions.

Chisholm imagines that Adam and Noah exist here in the actual world,w₁, and that each of them has his own collectionof properties. He assumes (followingGenesis 5) that Adamactually lives to the age of 930 years, whereas Noah lives to the ageof 950 years. Chisholm asks us to imagine a nearby possible world inwhich Adam lives one more year, dying at age 931 instead of 930, andin which Noah lives one year less, dying at age 949 instead of 950. Byimagining further similar one year alterations in life-span, wefinally reach a possible world in which Adam lives to the age of 950and Noah lives to the age of 930. Thus, in that world Adam and Noahhave “swapped ages”. In that world Adam has the age thatNoah has inw₁, and Noah has the age that Adam hasinw₁.

Chisholm then continues by imagining a series of worlds in which Adamand Noah gradually swap letters in their names. This series ends witha world in which they have “swapped names” altogether. Inthat world, Adam is called ‘Noah’, and Noah is called ‘Adam’.^[5] Going still further, Chisholm imagines a world,w_n, in which Adam and Noah have swappedall their qualitative properties, so that inw_n Adam has all the properties that Noahhas inw₁, and Noah has all the properties thatAdam has inw₁. Then Chisholm asks a series oftough questions:

Should we say of the Adam ofw_n that heis identical with the Noah ofw₁ and should we sayof the Noah ofw_n that he is identicalwith the Adam ofw₁? In other words, is there anx such thatx is Adam inw₁ andx is Noah inw_n, and is there ay such thaty is Noah inw₁ andy is Adam inw_n? And how are weto decide? … aren’t the two Adams, the two Noahs, and thetwo worlds indiscernible? Could God possibly have had a sufficientreason for creatingw₁ instead ofw_n? (3–4)

Chisholm extends the example. He points out that what has happened toAdam and Noah could happen to anyone. Let us say that a person’s“role” in a world is defined by the set of qualitativeproperties he has there. We can say that a possible world,w_s, is a “role swappingworld” of the actual world,w₁, just in casethe roles being played inw_s areprecisely the same as the roles being played inw₁, but at least some of those roles are beingplayed inw_s by someone other than theperson who plays the role inw₁. Role swappingworlds will therefore be indiscernible from each other and from theactual world except for the bare identities of the individuals playingthe roles. Chisholm says:

… there may be good ground for the existentialist’sangst; since, it would seem, God could have had no sufficientreason for choosing the world in which you play your present roleinstead of one in which you play mine. (4)

Chisholm’s remarks strongly suggest that he takes thesequestions to indicate that there is something deeply problematic aboutthe notion that the same individual can exist in different worlds withdifferent qualitative properties. Once we accept this notion, we seemto be on a path that will eventually lead us to the conclusion thatthere are indefinitely many worlds each of which is qualitativelyindiscernible from the real world, but differing from it only in theidentities of the individuals playing the various roles. Such aplurality of worlds seems undetectable and pointless.

In “Parts as Essential to Their Wholes” (1973), Chisholmpresents a structurally similar argument for a similar point. In thiscase, however, Chisholm’s target is a doctrine that he calls‘complete, unbridled mereological inessentialism’. This isthe view that nothing has its parts essentially — anythingthat’s made of parts could have been made of differentparts.

Chisholm starts his argument by sketching some outlandish examples,but then he turns to a familiar case. He asks us to consider a pair oftables,x andy. Each table is made of smallerparts. Just as he did in the case of Adam and Noah, Chisholm imaginesa series of small changes. Instead of gradual changes of properties,he here imagines gradual changes of parts. First, he describes apossible in world in whichx andy have exchangedone small part. Then he describes further possible worlds in whichmore small parts are exchanged. At last we reach a possible world inwhichx andy have exchanged all their parts:everything that is actually a part ofx has now become a partofy; and everything that is actually a part ofyhas now become a part ofx.

It’s not hard to see why Chisholm finds this example perplexing.For we have arrived at a world at which the table that is identifiedasx is composed of all and only the parts that in factcomposey; and the table that is identified asy iscomposed of all and only the parts that in fact composex. Isthis world in fact distinct from the actual world? What possible basisis there for saying that the table composed of thex parts isy? Why isn’t that tablex? It is partwiseindiscernible from the actualx.

Chisholm concludes this discussion by saying that “thesereflections, on the consequences of extreme mereologicalinessentialism, may suggest to us thatsome version ofmereological essentialism must be true…” (586). In theremaining pages of the paper Chisholm presents a view according towhich extreme mereological essentialism is true — but only forwhat he calls “primary objects”. Ordinary or“vulgar” objects such as tables and ships are not primaryobjects. Each such vulgar object may be understood to be a graduallyevolving sequence of primary objects. The vulgar object is constitutedby different primary objects at different times. Vulgar objects cangain and lose parts across times and possible worlds. Mereologicalessentialism is not true for them.

Chisholm strongly suggests that he thinks that vulgar objects occupy adecidedly lower rung on the ontological ladder. In a discussion of anexample involving a statue and a hunk of metal that temporarilyconstitutes it, he suggests that we might be content to say that thehunk of metal is a primary object and that this hunk of metaltemporarily has the property ofbeing statuesque. In thestrict and philosophical sense, it would be acceptable to say thatthere really is no such thing as “the statue”. Yet anotherpossible view would maintain that the statue is merely a mode of thehunk of metal. This too would diminish the ontological status of thestatue, though it would not banish the statue entirely. This latter“modal” view is explicitly defended in Chisholm’s1986 “Self-Profile” in [RMCp].

Chisholm discussed these problems in a number of papers and chapters.He did not always defend precisely the same combination of views, butcertain elements seemed to persist. He steadily maintained acommitment to mereological essentialism for ontologically fundamentalphysical objects; he tried in various ways to account for facts aboutpersistence, enumeration, and reidentification of ordinary objects byappeal to what he took to be ontologically more fundamental factsabout real “substances”. Chisholm’s work on thesetopics was remarkably provocative. Some have suggested that DavidLewis’s development of counterpart theory was intended (amongother things) to provide a metaphysical account of the structure ofpossible worlds that would avoid the strange implications thatChisholm described.

10. Ethics I — Intrinsic Value, Defeat, Theodicy

A reader starting out with [B&IV] for the first time might notanticipate how it is going to end. Indeed, even as the readerpatiently works his way through definitions, principles, remarks aboutwhat Brentano might have said, etc., the eventual point remains in thebackground. Nevertheless, as becomes clear in the final pages, thebook as a whole may be seen as one long argument for a distinctivesolution to the problem of evil.

Chisholm starts by explicating some of Brentano’s idiosyncraticmetaphysical views. Soon, however, Chisholm turns to some centralethical doctrines. Perhaps the first among these involves theintroduction of the concept ofcorrectness. In the firstinstance, we think of certain judgments as being correct. Thus, forexample, suppose it now seems to me that I am seeing something red. IfI judge that I am now seeing something red, it will be clear to methat my judgment is correct. If another person judges that I am notseeing something red, it will be clear to me that his judgment isincorrect. A similar thing could happen in the case of a judgment tothe effect that all squares are rectangles. If I make that judgmentwith full understanding of the concepts of squareness andrectangularity, it will be clear to me that my judgment is correct.Brentano evidently made use of this concept of correct judgment in anattempt to explain the notion of truth.

Chisholm claims that Brentano extended the application of the conceptof correctness so that it would apply toemotions as well. OnChisholm’s view, emotions are analogous to judgments in certainways. Corresponding to the notion of affirmation or “positivejudgment” we have the notion of love, or positive emotion.Corresponding to the notion of denial we have the notion of hatred, ornegative emotion. In addition, there is the concept of preference. IfI love one thing and hate another, then I prefer the one to the other.Similarly if I love one thing and am neutral about (neither loving norhating) the other, then I prefer the one to the other. And if I amneutral about one thing and hate the other, then I prefer the one tothe other. So we have the concept of preference as well.

Suppose I feel some pleasure. Suppose I love the fact that I amfeeling that pleasure. Suppose my love of this pleasure is intrinsic— I love the pleasure “in and for itself”. If Ireflect on this situation, I may recognize that my love is correct.This seems to mean, roughly, that I can see that it is appropriate, orfitting, for me to feel this strong pro-attitude toward thisexperience. The feeling of pleasure deserves, or merits, this sort ofpositive emotional reaction. If another person were to hate thatpleasure, or were to feel an equally strong anti-attitude toward it“in and for itself”, his hatred would be incorrect. We canjust see that a feeling of pleasure does not deserve this sort ofnegative emotional reaction.

Chisholm prefers to formulate these claims about the fittingness ofcertain emotional reactions to certain objects by appeal to afundamental concept in his ethics. This is the concept ofrequirement. Instead of saying that it is correct, orfitting, to have a certain emotional reaction to a certain object, wecan say that contemplation of that objectrequires thatemotional reaction. With these concepts at our disposal, Chisholmsuggests, we can define all the central concepts of ethics([B&IV], 53).

To say that one thingA isintrinsically better thananother thingB, according to Chisholm, is just to say that,for any person,x, contemplation of justA andB byx requires thatx preferA toB. Consider the fact that there are stones. While of coursewe might be glad that there are stones in virtue of their usefulness,it would not be correct to love this state of affairs for its ownsake. Mere contemplation of this state of affairs does not requireeither love or hate. It seems to be intrinsically neutral. Anintrinsically good thing is one that is intrinsically better than aneutral thing; an intrinsically bad thing is one such that anintrinsically neutral thing is intrinsically better than it. Thusaccording to Chisholm we may define intrinsic goodness, badness, andneutrality entirely in terms of preference and requirement.

In some cases a complex state of affairs contains some good parts andsome bad parts. We can assign positive numbers to the good parts andnegative numbers to the bad parts in order to represent theirrespective amounts of intrinsic value. The intrinsic value of thewhole state of affairs may just be the sum of the values of thoseparts. In such cases Chisholm would say that the positive value of thegood parts is “balanced off” by the negative value of thebad parts.

In other cases the interaction among the parts is not in this way justa matter of summation. Chisholm asks us to compare two cases. In Case1, I take pleasure in what I take to be Smith’s pleasure. InCase 2, I take an equal amount of pleasure in what I take to beJones’s pain. So here we are comparing two instances ofpleasure. Imagine that they are alike in what we may call the“raw amount” of pleasure involved. They differ in that oneis pleasure in something good whereas the other is pleasure insomething bad. Chisholm stipulates that the objects of these pleasures(Smith’s pleasure and Jones’s pain) are “intentionalobjects” — I take them to be happening, but I may bemistaken. Maybe they are just figments of my imagination.

Following Brentano (and Moore as well) Chisholm assumes that pleasurein the good is extra good, while pleasure in the bad is not so good.In these cases the value of the whole is either greater than, or lessthan, the sum of the values of the parts. (Since Jones’s pain isan intentional object of my pleasure in Case 2, its actual occurrenceis not entailed by the fact that I take pleasure in it; hence,according to Chisholm, it is not a “part” of the thatlarger state of affairs.) As a result, Case 1 is better than Case 2 inspite of the fact that Case 1 and Case 2 contain the same parts, withthe same values. The difference in value arises because of theappropriateness of the object of pleasure in Case 1, and thecorresponding inappropriateness of the object of the pleasure in Case2.

Case 2 illustrates what Chisholm calls “the defeat ofgood”. Case 2 contains a good part and no bad part; but thevalue of the whole is less than the value of the good part. Somehow,the value of the good part has been defeated without being balancedoff.

The defeat of evil would occur in a case in which some state ofaffairs contains a bad part that is worse than the whole, even thoughthe whole does not contain any good part that balances off the evil ofthe bad part. Consider Case 3 in which I take pain in what I take tobe my bad behavior. The badness of my pain may be mitigated by thefact that the object of this pain is something that requires that painbe taken in it — my own bad behavior. But it is possible that Iam mistaken about my bad behavior; maybe it never really happened.Thus, Case 3 does not contain that bad behavior as an actual part. Thestate of affairs may have no good part to balance off the badness ofmy pain, and yet it may be, as a whole, not as bad as its worst part.(It may be neutral; or even good.)

In the final five pages of [B&IV], Chisholm turns to theodicy. Heraises the question whether an omnipotent, omniscient, and benevolentdeity could create a world that contains some things that areintrinsically evil. Standard theistic answers are quickly rejected.(1) We cannot claim that evil is just an illusion, for in the case ofpain (for example) it is evident to the one who suffers it that it isgenuinely evil. Nor can we say (2) that some evils are required asmeans to good. For if God is omnipotent, he could find a way toproduce the good without making use of the evil means. (3) Some mightsay that God produces only “positive” states, such asexperiences of pleasure. The evil in the world occurs in“negative” states, and these are not due to God’screative efforts. But Chisholm points out that the negation of a goodstate is neutral, not bad. Furthermore, the actual existence of painis not a mere negation. It is as “positive” as theexistence of pleasure. (4) Some have claimed that God gave us freedom,which we then misused. Thus, it’s not God’s fault thatthere is evil in this world. But this “free will defense”is quickly rejected. The problem is that if God is omnipotent, hecould have created a different world in which we always freely chooseto act in the better way. Surely there must be such a world; surely itmust be better than the actual world in which we freely misbehave. Whydidn’t God actualize a world like that?

Chisholm then sketches his own solution to the problem of evil. Hesays ‘Some of the evil in the world is necessary for theenhancement of goodness. And the rest of the evil is defeated’([B&IV], 100). In a characteristically careful way, Chisholm doesnot assert that all the evil in this world is actually to be explainedin this way. He restricts himself to suggesting that Brentano wouldhave said such a thing, and that the conceptual machinery he hasdeveloped shows us how it is possible that the world could containsome evil even though the world was created by an omnipotent,omniscient, and benevolent god. Whether the evil in this world is infact defeated is another question. Chisholm concludes by saying‘The wise theodicist, I should think, would say that hedoesn’t know’ ([B&IV], 102).

11. Ethics II — Contrary to Duty Imperatives

One of Chisholm’s most influential critical papers in ethics is“Contrary-to-Duty Imperatives and Deontic Logic”. A numberof philosophers had noted the analogy between statements of necessity(e.g., ‘it’s necessary that red things arecolored.’) and statements of obligation (e.g., ‘it’sobligatory that promises are kept.’) They also noted thestructural similarities between the alethic modalities of necessity,possibility, and impossibility and the deontic modalities ofobligation, permission, and prohibition. It’s natural to thinkthat alethic necessity can be understood as truth at all possibleworlds. So, by analogy, some suggested systems of deontic logic inwhich obligation would be understood as truth at all morally perfectworlds. Statements of permission (and prohibition), would be taken tomean respectively truth at some (or no) morally perfect worlds. Thisgenerates a straightforward deontic logic.

Chisholm recognized, however, that any system based on thoseassumptions (and making use of the standard connectives) would beunable to express certain familiar and important facts. Among these isthe fact that in many cases our moral obligations depend essentiallyupon our moral failings. For example, a person might have a moralobligation to apologize for some misbehavior — yet obviously hewould not be apologizing in a morally perfect world, since in such aworld he would not have misbehaved in the first place. Presumably,there are no apologies in a morally perfect world. If we let‘O(Smith apologizes)’ express the idea that thereis a moral obligation for Smith to apologize, then the imaginedsemantics would apparently be unable to account for the possibilitythat the statement could be true. Smith does not apologize foranything at any morally perfect world.

Chisholm sketched a situation in which a person, (we can call him‘Smith’), had (a) an obligation to go to the aid of hisneighbor, (we can call him ‘Jones’). Assuming that itwould be best to notify Jones beforehand, Chisholm imagined that (b)Smith also had a conditional obligation to notify Jones in advance ifhe was going to come to his aid. On the other hand, (c) if he wasgoing to fail to come to Jones’s aid, then Smith’sobligation would be to avoid telling him that he was coming to hisaid. Finally, Chisholm assumed that (d) Smith was going to fail tocome to Jones’s aid.

There is nothing incoherent or even surprising about the situationChisholm described, yet there seemed to be no adequate way to expressthese statements in any extant system of deontic logic. No matter howthe obligation operators, material and formal conditionals, temporalrestrictions, etc. were juggled, the formal-language sentences simplydid not preserve the logical features of the familiar ordinarylanguage sentences. Consider, for example, the followingsentences:

(a1)O(Smith goes to the aid of Jones)

(b1)O(Smith goes to the aid of Jones → Smith tellsJones he is coming)

(c1) ~(Smith goes to the aid of Jones) →O(~Smith tellsJones he is coming)

(d1) ~(Smith goes to the aid of Jones).

Chisholm pointed out that in many extant systems of deontic logic,there is a general principle (“deontic detachment”) thatwould validate the inference from (a1) and (b1) to:

(e1)O(Smith tells Jones he is coming)

While (c1) and (d1) entail:

(f1)O(~Smith tells Jones he is coming)

The conjunction of (e1) and (f1) is a near contradiction, tellingSmith that he both has an obligation to warn Jones, and also that hehas an obligation not to warn Jones. If we assume that if you have anobligation to do something, then you don’t have an obligation toavoid doing it, we can derive an outright contradiction. Otherpossible representations of the situation fail to preserve otherlogical features of the original sentences.

Perhaps Chisholm was thinking that this puzzle could be solved byappeal to a system of deontic logic based on the notion ofrequirement. Maybe he thought that a good way to represent thesentences in question would be to say (a) that Smith had anon-overridden requirement to go to the aid of Jones; but that (b)that if Smith were going to go to the aid of Jones, then this factwould require that Smith notify Jones in advance; and furthermore (c)that if Smith were not going to go to the aid of Jones, then this factwould require that Smith not notify Jones in advance. Finally, thereis the stipulation (d) that Smith is not going to go to the aid ofJones. This solution to Chisholm’s puzzle would be of greatestinterest if it could be shown not only to provide a suitablesemantical interpretation for the four cited sentences, but also topave the way to a complete system of deontic logic adequate for theexpression of statements of conditional obligation.

Chisholm’s paper had a profound impact. It provoked renewedinterest in the logic of obligation. Several philosophers beganworking on new systems of deontic logic. Many of these new systemsincorporated a new idea — conditional obligation. This wasintended to provide a formal representation for the thought behindsuch statements as ‘if you don’t do what you ought to do,then you ought to apologize’.

12. Ethics III — The Ethics of Requirement

As we noted above in Section 10, Chisholm offered an account of theconcept of intrinsic goodness by appeal to the concept of requirement.In some places, Chisholm defended the idea that a thing isintrinsically good (bad) when contemplation of just that thingrequires that the contemplator loves (hates) it. But Chisholm wantedto go much further. In his discussion of these matters in [B&IV]Chisholm remarks:

This way of defining intrinsic value, then, makes use of the conceptofrequirement. And there is reason to think that the conceptof requirement is the central concept of ethics. It yields adequatedefinitions of the basic intrinsic value concepts. And it has thefollowing advantage as well: It provides a way of reducing theconcepts of the theory of value (“axiology”) to those ofethics (“deontology”). ([B&IV], 53)

In a couple of papers, Chisholm tried to show how the central conceptsof axiology and deontology can be defined by appeal to this onefundamental concept of requirement. These papers provide beautifulillustrations of one of Chisholm’s most characteristicphilosophical procedures — the clear definition of a tangle ofrelated concepts by appeal to a small set of undefined primitives. Inthis instance, he tries to explain confusing and problematic conceptsin the realm of ethics. He makes use of just a few undefined concepts.One of them —requirement — seems to carry all ofthe distinctively normative burden. (It would be interesting tocompare Chisholm’s efforts on behalf ofrequirement inethics with his efforts on behalf ofmore reasonable than inepistemology. The projects seem to illustrate the same methodologicalpropensity.)

In “The Ethics of Requirement”, Chisholm uses‘pRq’ to abbreviate ‘p wouldrequireq’. The relevant concept of requirement isoperative in the statement ‘If I were to promise to meet you forlunch that would require my meeting you for lunch.’ An actualrequirement forq comes into being whenq isrequired by something,p,and p actually occurs.Thus, if I actually did promise to meet you for lunch, then there is arequirement that I show up. But even if actual, the requirement isdefeasible. It may be overridden. When an action is actually requiredin this way, it counts at least as a prima facie obligation. If I meetan injured stranger on the way to lunch, and I alone am qualified totend to his injuries, then this further state of affairs overrides myprima facie obligation to meet you for lunch. In light of this,Chisholm defines all-in, or all things considered, obligation asnon-overridden requirement. In other words:

q is all things considered obligatory =_df
    ∃p(pRq &p occurs &~∃s(s occurs &~((p&s)Rq)))

This concept of obligation is not restricted to actions. It is aconcept of the “ought to be”. In order to explain the“ought to do”, Chisholm introduces another operator,‘A’. ‘Ap’ expresses the ideathat a personSsucceeds in bringing about the stateof affairsp. Making use this further expression, Chisholmdefines the ought to do by saying that the statement thatSought to bring about some state of affairsp just meansOAp.

Once we have a definition of all things considered obligation, it iseasy enough to define permission, forbiddenness, and gratuitousness.Chisholm does this in the expected ways.

A person has a prima facie obligation to do something when his doingit is required by something that happens. An all-in obligation ariseswhen there is a non-overridden requirement for him to do it. Chisholmclaims that this helps to solve the puzzle about conflicts ofobligation. Obviously, there can be conflicts of prima facieobligation. But there cannot be conflicts of all-in obligation.

In addition, Chisholm attempts to apply his conceptual scheme to someother long-standing puzzles in ethics. Among these is a puzzle aboutsupererogatory action. A supererogatory action is supposedly one thatis “beyond the call of duty” — something that wouldbe outstandingly good to do, but permissible to fail to do.Consequentialists have a hard time explaining how there could be anysuch things, since they typically say that duty requires us to do thebest we can. How can there be anything “beyond” that? Andhow could it be permissible to fail to do such a thing, if it is“extra good”?

Chisholm proposes to explain this by saying that a certain action issupererogatory when its “object” absolutely ought to be,but its agent is neither obligated to bring it about nor to refrainfrom bringing it about. He uses ‘PAq’ toabbreviate ‘S is permitted to bring aboutq’, where permission is understood to be the absence ofobligation to refrain:

S’s doingq is supererogatory=_dfOq &P~Aq &PAq

In subsequent papers, Chisholm made further claims about theusefulness of his concept of requirement.

13. Evolution and Criticism of Chisholm’s Views

Chisholm continued to work on a multitude of philosophical projectsover a period of many years. Though his central convictions remainedrelatively stable, he steadily revised the details of his formulationsin light of criticism from others and as a result of his ownreconsideration. Thus it is difficult to identify any chapter or paperas containing the final, official version of his view.

Nevertheless, there are several more or less fundamental Chisholmiandoctrines. We have sketched some of these doctrines. Here, we listsome things that Chisholm wrote in which he discussed the doctrines.We go on to mention some books and papers by others in which theyoffer exposition or criticism of Chisholm’s views.

In epistemology, Chisholm took as his fundamental primitive conceptthe idea we express when we say thatbelieving p is morereasonable for S than believing q would be. He tried to defineother terms of epistemic evaluation by appeal to this one.Furthermore, he assumed that we in fact do know many of the things wetake ourselves to know. He thought that a person can be justified inbelieving many foundational propositions about his own current mentalstate. He took one of the central projects of epistemology to be theformulation of epistemic principles that would show how we becomejustified in believing things about the external world, the past, andother problematic things on the basis of this present, internal,self-oriented foundation.

An early formulation of Chisholm’s view can be found in [PPS].Subsequent formulations were presented in the three editions of [TK]as well as in [FK]. [FK] is a collection of essays in epistemology.Some were newly written for [FK], but most were slightly revisedversions of articles that Chisholm had already published. [FK] is thusa comprehensive presentation of Chisholm’s distinctive views inepistemology.

Three excellent critical discussions are Heidelberger (1969), Foley(1997), and Sosa (1997). For comprehensive examinations ofChisholm’s epistemology, see Legum (2021 [Other InternetResources]) and the papers in [PLRMC].

One issue in Chisholm’s epistemological work that has beenwidely discussed is the problem of the criterion. Chisholm thoughtthat the problem arises because the fundamental problems ofepistemology were “What do we know?” and “What arethe criteria for knowledge?” and he argued that an answer toeither one required a prior answer to the other. He contended that theproblem could be addressed only by begging the question. He defendedthe common-sense view that we do know most of what common-sense sayswe know. For discussion, see Amico (1993), Lemos (2004), and McCain(2014 [Other Internet Resources]).

In metaphysics, Chisholm maintained that certain things are genuinelypersisting “substances”. Such things literally lastthrough time and survive changes in their properties. Each person issuch a being, retaining his or her strict identity over time. On theother hand, ordinary physical objects can be said to persist only inan extended sense. Since it has undergone changes of parts,today’s chair is only “loosely identical” withyesterday’s chair. Chisholm thought that a complex thing couldstrictly persist only if it retained all of its parts. Thus, heendorsed mereological essentialism. He thought that persons would haveto be ontologically simple in order for it to be possible for them tomaintain strict identity over time and through change. Thus, we arenot to be identified with our bodies.

An excellent presentation of Chisholm’s views on these matterscan be found in [P&O], as well as in his papers “Is There aMind-Body Problem?” (1978) and “On the Simplicity of theSoul” (1991). Chapters 4, 6, 7, and 13 of [OM] are especiallyhelpful. For a brief final discussion of the metaphysics of persons,see “Persons and Their Bodies: Some Unanswered Questions”(1996).

Plantinga (1975) contains a clear and helpful discussion ofChisholm’s views about mereological essentialism. For adrammatically non-Chisholmian view concerning the relation between aperson and that person’s body, see Baker (2002). Although he didnot explictly say that he was presenting an objection toChisholm’s view, Derek Parfit also defended a radicallydifferent approach to the nature of persons and the conditions for“personal identity” in Parfit (1971).

For a penetrating critical discussion of Chisholm’s metaphysicsof persons, see Quinn (1997).

Another metaphysical view that evidently fascinated Chisholm islibertarianism. According to this view, there is a special sort ofcausation in which an event is caused at least in part by an agent.When an agent thus “imminently” helps to bring about someeffect, his doing so is not determined by antecedent causalconditions. Thus, he freely makes something happen. This makes genuinemoral responsibility possible.

An early defense of this libertarianism can be found in “Freedomand Action” (1966). Another classic statement of the view can befound in Chisholm’s “He Could Have Done Otherwise”(1967a). Chisholm discussed these ideas in greater detail in ChapterII of [P&O] and then again in Chapters 1 and 2 of [OM]. Forfurther helpful work, including Chisholm’s “Human Freedomand the Self,” see Ekstrom (2000). See also Zimmerman (2010) andvon Wachter (2003).

Chisholm’s views in ontology underwent steady pruning. In anearlier period he believed in necessarily existent states of affairssome of which might occur and recur. He also believed in times, thoughnot in “concrete events”. Later he abandoned times andrecurrable states of affairs in favor of a sparser ontology including“one shot” states, but lacking times. The earlier view canbe found in “Events and Propositions” (1970), and“States of Affairs Again” (1971), and Chapter IV of[P&O]. The later view appears in “Events without Times: AnEssay in Ontology” (1990). See also Chapter 16 of [OM] andChapter 10 of [RTC].

Kim 1979 contains a very helpful critical exposition ofChisholm’s early view. Both views are discussed in Brandl 1997.See also Zimmerman (1997), Rosenkrantz (1998), and Steen (2008).

Chisholm was dubious about such things as sense data, appearances, the“looks” of things. He defended a different view about theontology of perception. An amusing early paper on this topic is“The Problem of the Speckled Hen” (1942). A more extendeddiscussion, including a presentation of Chisholm’s preferredtheory of appearing, can be found in Chapter 8 (“Sensing”)of [PPS]. More than forty years later, near the end of his career,Chisholm made some final remarks about appearances in Chapter 13 of[RTC].

Chisholm steadily thought there is something fundamentally right aboutBrentano’s Thesis — the idea that the psychologicalsomehow involves intentionality, and that in this respect it is unlikethe purely physical. Chisholm discussed this idea in his early paper“Sentences about Believing” (1955–6). He discussedit again in Chapter 11 (“Intentional Inexistence”) of[PPS]. A summary statement of the view can also be found inChisholm’s article “Intentionality” (1967c).Chisholm returned to this topic in Part IV (Chapters 10–14) of[OM].

Two excellent critical discussions of Brentano’s Thesis areSanford 1997 and Kim 2003.

The term ‘Chisholm’s Paradox’ has been credited toGraeme Forbes in his “Two Solutions to Chisholm’sParadox” (1984) The topic has been discussed extensively; see,for example, Salmon 1996. In his discussion of the mereologicalversion of the puzzle, Chisholm mentions Chandler 1966. Lewis 1986 isessential reading here. A good discussion of the whole problem oftransworld identity can be found in Mackie 2008.

One central Chisholmian view in ethics is the idea that all the mainconcepts of ethics can be defined by appeal to the concept ofrequirement. He defends this idea in his early paper“The Ethics of Requirement” (1964) and then again tenyears later in “Practical Reason and the Logic ofRequirement” (1974). The idea is briefly mentioned, but notdeveloped, in [B&IV] 53. See also Chisholm (2005).

For quite a long time Chisholm thought that the fundamental bearers ofintrinsic value are states of affairs. He also thought that in somecases, the value of a state of affairs could be defeated by thatstate’s occurrence within some larger context. In this he seemedto be working out some ideas of Brentano and Moore. One ofChisholm’s most moving presentations of this view can be foundin his Presidential Address before the Eastern Division of theAmerican Philosphical Association titled “The Defeat of Good andEvil” (1968). Many of his ideas about defeat, enhancement,mitigation, and organic unities are further developed in Chapters 6,7, and 8 of [B&IV]. For further discussion, see Lemos (2005),Gustafsson (2014), and Tucker (2016).

No list of Chisholm’s contributions to ethics could be completewithout mentioning “Contrary-to-Duty Imperatives and DeonticLogic” (1963).

Bibliography

Primary Literature

Cited Books by Chisholm

Cited Articles by Chisholm

• 1942, “The Problem of the Speckled Hen,”Mind51(204): 368–373.
• 1955–6, “Sentences about Believing,”Proceedings of the Aristotelian Society, 56:125–148.
• 1963, “Contrary-to-Duty Imperatives and DeonticLogic,”Analysis, 24: 33–36.
• 1964, “The Ethics of Requirement,”AmericanPhilosophical Quarterly, 1: 147–153.
• 1966, “Freedom and Action,” inFreedom andDeterminism, K. Lehrer (ed.), New York: Random House, pp.11–44; reprinted in Myles Brand (ed.),The Nature of HumanAction, Glenview, IL: Scott Foresman and Company, 1970, pp.283–292.
• 1967a, “He Could Have Done Otherwise,”Journal ofPhilosophy, 4: 409–417; reprinted in Myles Brand (ed.),The Nature of Human Action, Glenview, IL: Scott Foresman andCompany, 1970, pp. 293–301.
• 1967b, “Identity through Possible Worlds: SomeQuestions,”Noûs, 1(1): 1–8.
• 1967c, “Intentionality,”The Encyclopedia ofPhilosophy (Volume 5), Paul Edwards (ed.), London: Macmillan, pp.201–4.
• 1968, “The Defeat of Good and Evil,”,Proceedingsand Addresses of the American Philosophical Association, 42:21–38.
• 1970, “Events and Propositions,”Noûs,4: 15–24.
• 1971, “States of Affairs Again,”Noûs,5: 179–189.
• 1973, “Parts as Essential to Their Wholes,”Reviewof Metaphysics, 26: 581–603.
• 1974, “Practical Reason and the Logic of Requirement,”in S. Korner (ed.),Practical Reason, Oxford: BasilBlackwell, pp. 1–17.
• 1978, “Is There a Mind-Body Problem?,”Philosophical Exchange, 2: 24–34.
• 1986, “Self-Profile,” in [RMCp], pp. 3–77.
• 1990, “Events without Times: An Essay in Ontology,”Noûs, 24: 413–428.
• 1991, “On the Simplicity of the Soul,”Philosophical Perspectives, 5: 157–81.
• 1996, “Persons and Their Bodies: Some UnansweredQuestions,” in [RTC], pp. 99–105.
• 2005, “Intrinsic Value,” in ToniRønnow-Rasmussen & Michael J. Zimmerman (eds.),RecentWork on Intrinsic Value, Dordrecht: Springer, pp.1–10.

Secondary Literature

Collections

Cited Secondary Literature

• Amico, R., 1993,The Problem of the Criterion, Lanham,MD: Rowman and Littlefield.
• Baker, L. R., 2002, “The ontological status ofpersons,”Philosophy and Phenomenological Research,65(2): 370–88.
• Brandl, J., 1997, “Recurrent Problems — OnChisholm’s Two Theories of Events,” in [LLP], pp.457–477.
• Brentano, F., 1874,Psychology from an EmpiricalStandpoint, translated by A.C. Rancurello, D.B. Terrell, and L.McAlister, London: Routledge, 1973.
• Chandler, H., 1966, “Essence and Accident,”Analysis, 26: 185–188.
• Davidson, D., 1970, “Events as Particulars,”Noûs, 4(1): 25–32.
• Ekstrom, L. W. (ed.), 2000,Agency and Responsibility: Essayson the Metaphysics of Freedom, Boulder, CO: Westview.
• Foley, R., 1997, “Chisholm’s EpistemicPrinciples,” in [LLP], pp. 241–264.
• Forbes, G., 1984, “Two Solutions to Chisholm’sParadox,”Philosophical Studies, 46:171–187.
• Gustafsson, J. F., 2014, “Neither ‘Good’ inTerms of ‘Better’ nor ‘Better’ in Terms of‘Good’,”Noûs, 48(1):466–473.
• Heidelberger, H., 1969, “Chisholm’s EpistemicPrinciples,”Noûs, 3(1): 73–82.
• Kim, J., 1979, “States of Affairs, Events, andPropositions,” in [EPRMC], pp. 147–162.
• Kim, Jaegwon, 2003, “Chisholm’s Legacy onIntentionality,” in [PLRMC], pp. 649–662.
• Lemos, N., 2004,Common Sense: A Contemporary Defense,New York: Cambridge University Press.
• –––, 2005, “Chisholm’s Definition ofOrganic Unity,” in Toni Rønnow-Rasmussen & Michael J.Zimmerman (eds.),Recent Work on Intrinsic Value, Dordrecht:Springer, pp. 319–23.
• Lewis, D., 1986,On the Plurality of Worlds, Oxford:Blackwell.
• Mackie, P., 2008, “Transworld Identity,”TheStanford Encyclopedia of Philosophy (Fall 2008 Edition), EdwardN. Zalta (ed.), URL = <https://plato.stanford.edu/archives/fall2008/entries/identity-transworld/>.
• Parfit, D., 1971, “Personal Identity,”ThePhilosophical Review, 80(1): 3–27.
• Plantinga, A., 1975, “On Mereological Essentialism,”Review of Metaphysics, 28(3): 468–476.
• Quinn, P., 1997, “Tiny Selves: Chisholm on the Simplicity ofthe Soul,” in [LLP], pp. 56–66.
• Rosenkrantz, G. S., 1998, “A Realistic Theory of Categories:An Essay in Ontology,”The Philosophical Review,107(4): 650–3.
• Salmon, N., 1996, “Trans-world Identification andStipulation,”Philosophical Studies, 84(2):203–223.
• Sanford, D., 1997, “Chisholm on Brentano’sThesis,” in [LLP], pp. 201–214.
• Sosa, E., 1997, “Chisholm’s Epistemology and EpistemicInternalism,” in [LLP], pp. 267–287.
• Steen, M., 2008, “Chisholm’s chaning conception ofordinary objects,”Grazier Philosophische Studien,76(1): 15–46
• Taylor, R., 1975, “A Tribute,” in [A&M], pp.1–8.
• Tucker, M., 2016, “Two Kinds of Value Pluralism,”Utilitas, 28(3): 333–46.
• von Wachter, D., 2003, “Free Agents as Cause,” inKlaus Petrus (ed.),On Human Persons, Frankfurt: OntosVerlag, pp. 183–94.
• Zimmerman, D., 1997, “Chisholm and the Essences ofEvents,” in [LLP], pp. 73–93.
• Zimmerman, M. J., 2010, “Chisholm,” in TimothyO’Connor & Constantine Sandis (eds.),A Companion to thePhilosophy of Action, Oxford: Wiley-Blackwell, pp.578–588.

Academic Tools

	How to cite this entry.
	Preview the PDF version of this entry at theFriends of the SEP Society.
	Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO).
	Enhanced bibliography for this entryatPhilPapers, with links to its database.

Other Internet Resources

• Legum, R. 2021, “>Roderick M. Chisholm: Epistemology,” entry in theInternet Encyclopedia of Philosophy.
• McCain, K., 2014, “Problem of the Criterion,” entry in theInternet Encyclopedia of Philosophy.
• Banquet Address for the Chisholm Memorial Conference, by Fred Feldman.
• The Philosophical Lexicon

Related Entries

Brentano, Franz |categories |epistemology |epistemology: naturalism in |events |free will |identity: over time |identity: transworld |intentionality |logic: deontic |perception: the problem of |self-knowledge |sense data |value: intrinsic vs. extrinsic