Generally regarded as one of the most important philosophers to writein English, David Hume (1711–1776) was also well known in hisown time as an historian and essayist. A master stylist in any genre,his major philosophical works—A Treatise of HumanNature (1739–1740), theEnquiries concerning HumanUnderstanding (1748) andconcerning the Principles ofMorals (1751), as well as his posthumously publishedDialogues concerning Natural Religion (1779)—remainwidely and deeply influential.

Although Hume’s more conservative contemporaries denounced hiswritings as works of scepticism and atheism, his influence is evidentin the moral philosophy and economic writings of his close friend AdamSmith. Kant reported that Hume’s work woke him from his“dogmatic slumbers” (Prolegomena, Introduction)and Jeremy Bentham remarked that reading Hume “caused the scalesto fall” from his eyes (“A Fragment on Government”,chapter 1, paragraph 36, footnote 2). Charles Darwin regarded hiswork as a central influence on the theory of evolution. The diversedirections in which these writers took what they gleaned from readinghim reflect both the richness of their sources and the wide range ofhis empiricism. Today, philosophers recognize Hume as a thoroughgoingexponent of philosophical naturalism, as a precursor of contemporarycognitive science, and as the inspiration for several of the mostsignificant types of ethical theory developed in contemporary moralphilosophy.

• 1. Life and Works
• 2. The relation between theTreatise and theEnquiries
• 3. Philosophical Project
• 4. Account of the Mind
  • 4.1 The Copy Principle
  • 4.2 Hume’s Account of Definition
  • 4.3 Principles of Association
• 5. Causation
  • 5.1 Causal Inference: Critical Phase
  • 5.2 Causal Inference: Constructive Phase
  • 5.3 Belief
• 6. The Idea of Necessary Connection
  • 6.1 Necessary Connection: Critical Phase
  • 6.2 Necessary Connection: Constructive Phase
• 7. Moral Philosophy
  • 7.1 Moral Rationalism: Critical Phase in theTreatise
  • 7.2 Sentimentalism: Constructive Phase
  • 7.3 Self-Interest Theories: Critical Phase in theEnquiry
  • 7.4 Justice: Constructive Phase
• 8. Philosophy of Religion
  • 8.1 The Characters
  • 8.2 God’s Natural Attributes
  • 8.3 God’s Moral Attributes
  • 8.4 The Conclusion: Part 12
• Bibliography
  • Primary Literature
  • Secondary Literature
• Academic Tools
• Other Internet Resources
• Related Entries

1. Life and Works

Born in Edinburgh, Hume spent his childhood at Ninewells, hisfamily’s modest estate in the border lowlands. He came from a“good family” (MOL 2)—socially well connected butnot wealthy. His father died just after David’s second birthday,leaving him and his elder brother and sister in

the care of our Mother, a woman of singular Merit, who, though youngand handsome, devoted herself entirely to the rearing and educating ofher Children. (MOL 3)

Katherine Falconer Hume realized that David was uncommonly precocious,so when his older brother went up to Edinburgh University, Hume wentwith him, although he was only 10 or 11. There he studied Latin andGreek, read widely in history and literature, ancient and modernphilosophy, and also did some mathematics and naturalphilosophy—what we now call natural science.

The education David received, both at home and at the university,aimed at training pupils to a life of virtue regulated by sternScottish Calvinist strictures. Prayers and sermons were prominentaspects of his home and university life. At some point, Hume readThe Whole Duty of Man, a widely circulated Anglicandevotional tract that details our duties to God, our fellow humanbeings, and ourselves.

Hume’s family thought him suited for a legal career, but hefound the law “nauseous”, preferring to read classicaltexts, especially Cicero. He decided to become a “Scholar andPhilosopher”, and followed a rigorous program of reading andreflection for three years until “there seem’d to beopen’d up to me a new Scene of Thought” (HL 3.2). Theintensity of developing his philosophical vision precipitated apsychological crisis in the isolated scholar.

The crisis eventually passed, and Hume remained intent on articulatinghis “new Scene of Thought”. As a second son, hisinheritance was meager, so he moved to France, where he could livecheaply, and finally settled in La Flèche, a sleepy village inAnjou best known for its Jesuit college where Descartes and Mersennehad studied a century before. Here he read French and othercontinental authors, especially Malebranche, Dubos, and Bayle, andoccasionally baited the Jesuits with arguments attacking theirbeliefs. By this time, Hume had not only rejected the religiousbeliefs with which he was raised, but was also opposed to organizedreligion in general, an opposition that remained constant throughouthis life. In 1734, when he was only 23, he began writingATreatise of Human Nature.

Hume returned to England in 1737 to ready theTreatise forthe press. To curry favor with Joseph Butler (1692–1752), he“castrated” his manuscript, deleting his controversialdiscussion of miracles, along with other “nobler parts”(HL 6.2). Book I, “Of the Understanding”, and Book II,“Of the Passions”, appeared anonymously in 1739. The nextyear saw the publication of Book III, “Of Morals”, as wellas his anonymous “Abstract” of Books I and II.

TheTreatise was no literary sensation, but it didn’tfall “deadborn from the press” (MOL 6), as Humedisappointedly described its reception. Despite his surgicaldeletions, it attracted enough of “a Murmour among theZealots” (MOL 6) to fuel his lifelong reputation as an atheistand a sceptic. When he applied for the Chair of Ethics and Pneumatical(“Mental”) Philosophy at Edinburgh in 1745, his reputationprovoked vocal and ultimately successful opposition. Six years later,he stood for the Chair of Logic at Glasgow, only to be turned downagain. Hume never held an academic post.

In 1745, he accepted a position as a young nobleman’s tutor,only to discover that his charge was insane. A year later he becamesecretary to his cousin, Lieutenant General James St Clair, eventuallyaccompanying him on an extended diplomatic mission in Austria andItaly.

In 1748,An Enquiry concerning Human Understanding appeared,covering the central ideas of Book I of theTreatise and hisdiscussion of liberty and necessity from Book II. He also includedmaterial he had excised from theTreatise. In 1751, hepublishedAn Enquiry concerning the Principles of Morals, a“recasting” of Book III of theTreatise, which hedescribed as “incomparably the best” of all his work (MOL10). More essays, thePolitical Discourses, appeared in 1752,and Hume’s correspondence reveals that a draft of theDialogues concerning Natural Religion was also underway atthis time.

An offer to serve as Librarian to the Edinburgh Faculty of Advocatesgave Hume the opportunity to begin another project, aHistory ofEngland, using the law library’s excellent resources.Published in six volumes between 1754 and 1762, hisHistorywas a bestseller well into the next century, giving him the financialindependence he had long sought. But even as a librarian, Hume’sreputation as an atheist and sceptic dogged him. One of his orders for“indecent Books” prompted an unsuccessful move for hisdismissal and excommunication from the Kirk. Friends and publisherspersuaded him to suppress some of his more controversial writings onreligion during his lifetime.

In 1763, Hume accepted a position as private secretary to the BritishAmbassador to France. During his three-year stay in Paris, he becameSecretary to the Embassy, and eventually itschargéd’affaires. He became the rage of the Parisian salons,enjoying the conversation and company of famous Europeanintellectuals. He was known for his love of good food and wine, aswell as his enjoyment of the attentions and affections of women.

Hume returned to Edinburgh in 1769. He built a house inEdinburgh’s New Town, and spent his autumnal years quietly andcomfortably, dining and conversing with friends, not all of whom were“studious and literary”, for he also found that his“company was not unacceptable to the young and careless”(MOL 21). He spent considerable time revising his works for neweditions of hisEssays and Treatises, which contained hiscollectedEssays, the twoEnquiries,ADissertation on the Passions, andThe Natural History ofReligion, but—significantly—notA Treatise ofHuman Nature.

In 1775, Hume was diagnosed with intestinal cancer. Before his deathin 1776, he arranged for the posthumous publication of his mostcontroversial work, theDialogues concerning NaturalReligion, and composed a brief autobiography, “My OwnLife”. Although there was much curiosity about how “thegreat infidel” would face his death, his friends agreed that heprepared himself with the same peaceful cheer that characterized hislife.

2. The relation between theTreatise and theEnquiries

In 1775, as he was readying a revised edition of hisEssays andTreatises for the press, Hume sent his publisher an“Advertisement”, asking that it be included in this andany subsequent edition of his works. In it, he complains that hiscritics focused “all their batteries” on theTreatise, “that juvenile work”, which hepublished anonymously and never acknowledged. He urges his readers toregard theEnquiries “as containing his philosophicalsentiments and principles”, assuring his publisher that theyprovide “a compleat answer” to his critics.

Hume’s apparent disavowal of theTreatise and hisregard for theEnquiries raise a question about how we shouldread his work. Should we take his statements literally and let theEnquiries represent his considered view, or should we ignorehis “Advertisement” and take theTreatise as thebest statement of his position?

Both options presuppose that the differences between theTreatise and theEnquiries are substantial enough towarrant taking one or the other as best representing Hume’sviews, but there are good reasons for doubting this. Even in the“Advertisement”, Hume says, “Most of the principles,and reasonings, contained in this volume, were published” in theTreatise. He repeats his conviction that he was guilty of“going to press too early”, and that his aim in theEnquiries was to “cast the whole anew … wheresome negligences in his former reasoning and more in the expression,are … corrected”.

Hume’s description of his aims suggests another option. Ratherthan repudiating theTreatise, perhaps his recasting of itrepresents a shift in the way he presents his “principles andreasoning” rather than a substantive change in what he has tosay. He reinforces this option when he says of the firstEnquiry that the “philosophical Principles are the samein both” and that “By shortening & simplifying theQuestions, I really render them much more complete” (HL 73.2).He also comments in “My Own Life” that theTreatise’s lack of success “proceeded more fromthe manner than the matter”—more from itsstructure than its content (MOL 8). It is not unreasonable toconclude that Hume’s recasting of theTreatise wasdesigned to address this issue, which suggests that we mightunderstand him best by reading both works, despite their differences,together.

3. Philosophical Project

As the title of theTreatise proclaims, Hume’s subjectis human nature. He summarizes his project in its subtitle: “anattempt to introduce the experimental method into moralsubjects”. In his day, “moral” meant anythingconcerned with human nature, not just ethics, as he makes clear at thebeginning of the firstEnquiry, where he defines “moralphilosophy” as “the science of human nature” (EHU1.1/5). Hume’s aim is to bring the scientific method to bear onthe study of human nature.

Hume’s early studies of philosophical “systems”convinced him that philosophy was in a sorry state and in dire need ofreform. When he was only 18 years old, he complained in a letter thatanyone familiar with philosophy realizes that it is embroiled in“endless Disputes” (HL 3.2). The ancient philosophers, onwhom he had been concentrating, replicated the errors their naturalphilosophers made. They advanced theories that were “entirelyHypothetical”, depending “more upon Invention thanExperience”. He objects that they consulted their imagination inconstructing their views about virtue and happiness, “withoutregarding human Nature, upon which every moral Conclusion mustdepend”. The youthful Hume resolved to avoid these mistakes inhis own work, by making human nature his “principal Study, &the Source from which I would derive every Truth” (HL 3.6).

Even at this early stage, the roots of Hume’s mature approach tothe reform of philosophy are evident. He was convinced that the onlyway to improve philosophy was to make the investigation of humannature central—and empirical (HL 3.2). The problem with ancientphilosophy was its reliance on “hypotheses”—claimsbased on speculation and invention rather than experience andobservation.

By the time Hume began to write theTreatise three yearslater, he had immersed himself in the works of the modernphilosophers, but found them disturbing, not least because they madethe same mistakes the ancients did, while professing to avoid them.Why, Hume asks, haven’t philosophers been able to make thespectacular progress in understanding human nature that naturalphilosophers—whom we now call“scientists”—have recently achieved in the physicalsciences? His answer is that while scientists have cured themselves oftheir “passion for hypotheses and systems”, philosophershaven’t yet purged themselves of this temptation. Their theorieswere too speculative, relied ona priori assumptions, andpaid too little attention to what human nature is actually like.Instead of helping us understand ourselves, modern philosophers weremired in interminable disputes—evident even to “the rabblewithout doors”—giving rise to “the common prejudiceagainst metaphysical reasonings of all kinds”, that is,“every kind of argument which is in any way abstruse, andrequires some attention to be comprehended” (T xiv.3).

To make progress, Hume maintains, we need to “reject everysystem … however subtle or ingenious, which is not founded onfact and observation”. These systems, covering a wide range ofentrenched and influential metaphysical and theological views, purportto have discovered principles that give us a deeper and more certainknowledge of ultimate reality. But Hume argues that in attempting togo beyond anything we can possibly experience, these metaphysicaltheories try to “penetrate into subjects utterly inaccessible tothe understanding” (EHU 1.11/11), which makes their claims tohave found the “ultimate principles” of human nature notjust false, but unintelligible. These “airy sciences”, asHume calls them, have only the “air” of science (EHU1.12/12).

Worse still, these metaphysical systems are smokescreens for“popular superstitions” that attempt to overwhelm us withreligious fears and prejudices (EHU 1.11/11). Hume has in mind avariety of doctrines that need metaphysical cover to lookrespectable—arguments for the existence of God, the immortalityof the soul, and the nature of God’s particular providence.Metaphysics aids and abets these and other superstitiousdoctrines.

But he insists that because these metaphysical and theological systemsare objectionable, it doesn’t mean we should give up doingphilosophy. Instead, we need to appreciate “the necessity ofcarrying the war into the most secret recesses of the enemy”.The only way to resist the allure of these pseudo–sciences is toengage with them, countering their “abstruse …metaphysical jargon” with “accurate and justreasoning” (EHU 1.12/12).

This means that the initial phase of Hume’s project must becritical. A prominent part of this aspect of his project isto “discover the proper province of humanreason”—determining the extent and limits ofreason’s powers and capacities (EHU 1.12/12). He believes thathis investigation will show that metaphysics as the quest forunderstanding the ultimate nature of reality is beyond reason’sscope.

Scholars once emphasized thiscritical phase at the expenseof the rest of Hume’s project, encouraging the charge that hewas just a negative skeptic, who rejects the views of others withoutdefending any positive position himself. But while he is indeedskeptical about the possibility of metaphysical insights that godeeper than science can, investigating the proper province of reasonisn’t only a critical activity. His critique of metaphysicsclears the way for theconstructive phase of hisproject—the development of an empirical science of humannature—and Hume is not at all skeptical about its prospects.

In his “Introduction” to theTreatise, Humelaunches the constructive phase of his project by proposing nothingless than “a compleat system of the sciences, built on afoundation entirely new” (T xvi.6). The new foundation is thescientific study of human nature. He argues that all the sciences havesome relation to human nature, “even Mathematics, NaturalPhilosophy, and Natural Religion” (T xv.4). They are all humanactivities, so what we are able to accomplish in them depends onunderstanding what kinds of questions we are able to handle and whatsorts we must leave alone. If we have a better grasp of the scope andlimits of our understanding, the nature of our ideas, and theoperations we perform in reasoning about them, there is no tellingwhat improvements we might make in these sciences.

We should expect even more improvement in the sciences that are moreclosely connected to the study of human nature: “Logic,Morals, Criticism, and Politics”. Many longstandingphilosophical debates are about the nature of ourideas—causation, liberty, virtue and beauty—so gettingclear about their content should help us cut through these“endless disputes”.

As the science of human nature is the only solid foundation for theother sciences, “the only solid foundation we can give to thisscience itself must be laid on experience and observation” (Txvi.7). Although Hume does not mention him by name, Newton(1642–1727) is his hero. He accepts the Newtonian maxim“Hypotheses non fingo”, roughly, “I do notdo hypotheses”. Any laws we discover must be established byobservation and experiment.

Hume is proposing anempiricist alternative to traditionala priori metaphysics. His empiricism isnaturalisticin that it refuses to countenance any appeal to thesupernatural in the explanation of human nature. As anaturalist, he aims to account for the way our minds work in a mannerthat is consistent with a Newtonian picture of the world.

Hume portrays his scientific study of human nature as a kind ofmental geography oranatomy of the mind (EHU1.13/13; T 2.1.12.2/326). In the first section of the firstEnquiry, he says that it has two principal tasks, one purelydescriptive, the otherexplanatory. Mental geographyconsists in delineating “the distinct parts and powers” ofthe mind (EHU 1.13/3). While everyone can make some sense of the basicdistinctions among the mind’s contents and operations, morefine–grained distinctions are harder to grasp.

Hume, however, wants to go much further. He wants toexplainhow the mind works by discovering its “secret springs andprinciples”. He reminds us that astronomers, for a long time,were content with proving the “motions, order, and magnitude ofthe heavenly bodies”. But then “aphilosopher”—Newton—went beyond them and determined“the laws and forces, by which the revolutions of the planetsare governed and directed” (EHU 1.15/14). Newton’s exampleled other natural philosophers to similar explanatory successes. Humebelieves he will be equally successful in finding the fundamental lawsgoverning our “mental powers and economy”, if he followsthe same caution Newton exhibited in carrying out his inquiries.

Newton’s scientific method provides Hume with a template forintroducing the experimental method into his investigation of themind. InAn Enquiry concerning the Principles of Morals, hesays he will follow “a very simple method” that hebelieves will bring about a transformation in the study of humannature. Following Newton’s example, he argues that we should“reject every system … however subtile or ingenious,which is not founded on fact and observation”, and accept onlyarguments derived from experience. When we inquire about human nature,since we are asking “a question of fact, not of abstractscience”, we must rely on experience and observation (EPM1.10/173–174).

As the fledgling Newton of the moral sciences, Hume wants to find aset of laws that explain how the mind’scontents—perceptions, as he calls them—come andgo in the mind and how simple perceptions combine to form complexperceptions in ways that explain human thought, belief, feeling andaction.

Newton’s achievement was that he was able to explain diverse andcomplex physical phenomena in terms of a few general principles. Likehim, Hume proposes to explain “all effects from the simplest andfewest causes” (T xvii.8). He predicts that it is likely thatone “principle of the mind depends on another” and thatthis principle may in turn be brought under another principle even“more general and universal” (EHU 1.15/15). But heemphasizes that while he will try to find the most general principles,rendering them as universal as possible, all of his explanations mustbe based completely on experience.

Although philosophy, as an empirical enterprise, is itself bound byexperience, this is not a defect in the science of human nature. Thesame is true for all the sciences: “None of them can go beyondexperience, or establish any principles which are not founded on thatauthority” (T Intro 10). Explanations must come to an endsomewhere. When we see that we have “arriv’d at the utmostextent of human reason, we sit down contented”, for the onlyreason we can give for our most general principles is “ourexperience of their reality” (T 9).

Hume is Newtonian in much more than method. He sees that Newton issignificantly different from John Locke (1632–1704) and theother Royal Society natural philosophers, because he rejects theirmechanist picture of the world. Newton’s greatest discovery, theLaw of Gravitation, is not a mechanical law. Hume explicitly modelshis account of the fundamental principles of the mind’soperations—the principles of association—on the idea ofgravitational attraction. By appealing to these same principlesthroughout, Hume gives an explanation of these diverse phenomena thatenable him to provide a unified and economical account of themind.

4. Account of the Mind

To explain the workings of our minds with the economy Newton displayedin his physics, Hume introduces the minimal amount of machinery hethinks is necessary to account for the mind’s operations. Eachpiece is warranted by experience.

The early modern period was the heyday of the investigation of theideas of causation, moral good and evil, and many otherphilosophically contested ideas. Every modern philosopher acceptedsome version of thetheory of ideas—the view that weimmediately perceive certain mental entities calledideas,but don’t have direct access to physical objects. Hume holds anempiricist version of the theory, because he thinks thateverything we believe is ultimately traceable to experience.

He begins with an account ofperceptions, because he believesthat any intelligible philosophical question must be asked andanswered in those terms. He usesperception to designate anymental content whatsoever, and divides perceptions into twocategories,impressions andideas.

Impressions includesensations as well asdesires,passions, andemotions.Ideas are “the faint images of these in thinking andreasoning” (T 1.1.1.1/1). He thinks everyone will recognize hisdistinction, since everyone is aware of the difference betweenfeeling andthinking. It is the difference betweenfeeling the pain of your present sunburn andrecalling last year’s sunburn.

Hume distinguishes two kinds of impressions:impressions ofsensation, ororiginal impressions, andimpressionsof reflection, orsecondary impressions. Impressions ofsensation include the feelings we get from our five senses as well aspains and pleasures, all of which arise in us “originally, fromunknown causes” (T 1.1.2.1/7). He calls themoriginalbecause trying to determine their ultimate causes would take us beyondanything we can experience. Any intelligible investigation must stopwith them.

Impressions of reflection include desires, emotions, passions, andsentiments. They are essentially reactions or responses to ideas,which is why he calls themsecondary. Your memories of lastyear’s sunburn are ideas, copies of the original impressions youhad when the sunburn occurred. Recalling those ideas causes you tofear that you’ll get another sunburn this year, tohope that you won’t, and towant to takeproper precautions to avoid overexposure to the sun.

Perceptions—both impressions and ideas—may be eithersimple orcomplex. Complex impressions are made upof a group of simple impressions. My impression of the violet I justpicked is complex. Among the ways it affects my senses are itsbrilliant purple color and its sweet smell. I can separate anddistinguish its color and smell from the rest of my impressions of theviolet. Its color and smell are simple impressions, which can’tbe broken down further because they have no component parts.

Hume initially distinguishes impressions and ideas in terms of theirdegree offorce andvivacity. Impressions are moreforceful and vivacious than ideas. My impression of this ripetomato’s bright red color is as vivid as anything could be. Lastyear’s tomatoes were just as vivid when I was looking at them,but now my idea of them is much less vivid than my impressions of thetomato in front of me. Since last year’s tomatoes were the samecolor, the difference can’t be that they are different shades ofred; the difference must lie in the sharpness, clarity, and brightnessof my impressions—theirforce andvivacity. Atvarious times, Hume tries other ways of characterizing the differencebetween impressions and ideas, but he was never completely satisfiedwith them. Still, what he says works well enough to give us a handleon the felt differences between impressions and ideas.

When Hume distinguishes impressions and ideas in terms of theirrelative force and vivacity, he is pointing out something that isgenerally true of them as a matter of fact. On occasion, in dreams ora high fever, ideas may approach the force and vivacity ofimpressions, but these are exceptions that provethe—empirical—rule. In general, impressions and ideas areso different that no one can deny the distinction.

Although nothing seems freer than the power of thought, whichisn’t “restrained within the limits of nature andreality” (EHU 2.4/18), Hume insists that our imagination is infact “confined within very narrow limits”. We can separateand combine our ideas in new and even bizarre ways, imaginingcreatures we’ve never seen or faraway galaxies, but all thematerials of thinking are ultimately derived from our impressions.Since “all our ideas or more feeble perceptions are copies ofour impressions or more lively ones”; we are restricted to“compounding, transporting, augmenting, or diminishing thematerials afforded us by the senses and experience” (EHU2.5/19).

4.1 The Copy Principle

In theTreatise, Hume qualifies his claim that our ideas arecopies of our impressions, making clear that it applies only to therelation betweensimple ideas andsimpleimpressions. He offers this “general proposition”,usually called theCopy Principle, as his “firstprinciple … in the science of human nature”:

All our simple ideas in their first appearance are deriv’d fromsimple impressions, which are correspondent to them, and which theyexactly represent. (T 1.1.1.7/4)

He presents the principle as something that everyone’sexperience confirms, but he also gives an argument to establishit.

He argues first that there is a one–to–one correspondencebetween simple ideas and simple impressions. He can’tprove that this correspondence holds universally, since hecan’t examine every individual impression and idea. But he is soconfident the correspondence holds that he challenges anyone whodoubts it to produce an example of a simple impression without acorresponding simple idea, or a simple idea without a correspondingsimple impression. Since he is certain they will fail, he concludesthat there is aconstant conjunction between simpleimpressions and simple ideas.

Next, he maintains that this constant conjunction is so universal thatthe correspondence can’t be a matter of chance. There must be acausal connection between them, but do ideas cause impressions or doimpressions cause ideas?

Finally, he argues that experience tells us that simple impressionsalways precede and thus cause their corresponding ideas. To supportthis claim, he appeals to two sorts of cases. First, if you want togive a child an idea of the taste of pineapple, you give her a pieceof pineapple to eat. When you do, you are giving her an impression ofthe pineapple’s taste. You never go the other way round. Hisother case involves a person born blind, who won’t have ideas ofcolor because he won’t have impressions of color.

The Copy Principle is an empirical thesis, which he emphasizes byoffering “one contradictory phenomenon” as an empiricalcounterexample to the principle. He imagines someone who has had thesame sorts of experiences of colors most of us have had, but has neverexperienced a certain shade of blue. Hume thinks that if he orders allthe shades of blue he has experienced from the darkest to thelightest, he will see immediately that there is a gap where themissing shade should be. Then he asks

Whether ‘tis possible for him, from his own imagination, to… raise up to himself the idea of that particular shade,tho’ it had never been convey’d to him by his senses? Ibelieve there are few but will be of opinion that he can; and this mayserve as a proof, that the simple ideas are not always deriv’dfrom the correspondent impressions; tho’ the instance is soparticular and singular, that ‘tis scarce worth our observing,and does not merit that for it alone we shou’d alter our generalmaxim. (T 1.1.1.10/6)

Hume repeats the case of the missing shade almost verbatim in thefirstEnquiry. While scholars have wondered exactly how theperson might supply the missing shade, he seems unconcerned with thedetails. For Hume, once again the exception provesthe—empirical—rule.

4.2 Hume’s Account of Definition

Although Hume’s distinctive brand of empiricism is oftenidentified with his commitment to the Copy Principle, his use of theprinciple’s reverse in his account of definition is perhaps themore innovative element of his system.

As his diagnosis of traditional metaphysics reveals, Hume believesthat

the chief obstacle … to our improvement in the moral ormetaphysical sciences is the obscurity of the ideas, and ambiguity ofthe terms. (EHU 7.1.2/61)

Conventional definitions—replacing terms with theirsynonyms—merely replicate philosophical confusions and neverbreak out of a narrow definitional circle. Getting clear about thecontent of the ideas and the meanings of the terms we areinvestigating requires something else.

Hume argues that we must “pass from words to the true and realsubject of the controversy”—ideas. He believes he hasfound a way to accurately determine their content—his account ofdefinition. He touts it as “a new microscope or species ofoptics”, predicting that it will produce equally dramaticresults in the moral sciences as its hardwarecounterparts—telescopes and microscopes—have produced innatural philosophy (EHU 7.1.4/62).

Hume’s account of definition uses a simple series of tests todetermine cognitive content. Begin with a term. Ask what idea isannexed to it. If there is no such idea, then the term has nocognitive content, however prominently it figures in philosophy ortheology. If there is an idea annexed to the term, and it is complex,break it down into the simple ideas that compose it, and trace themback to their original impressions. If the process fails at any point,the idea in question lacks cognitive content. When carried throughsuccessfully, however, it yields a “justdefinition”—a precise account of the troublesomeidea’s content.

Hume uses his account of definition in thecritical phaseofhis project to show that many of the central concepts of traditionalmetaphysics lack intelligible content. He also uses it in theconstructive phase to determine the exact meaning of ourterms and ideas.

4.3 Principles of Association

Although we are capable of separating and combining our simple ideasas we please, there is, nevertheless, a regular order to our thoughts.If ideas occurred to us completely randomly, so that all our thoughtswere “loose and unconnected”, we wouldn’t be able tothink coherently (T 1.1.4.1/10). This suggests that

There is a secret tie or union among particular ideas, which causesthe mind to conjoin them more frequently, and makes the one, upon itsappearance, introduce the other. (Abstract 35)

Hume explains this “tie or union” in terms of themind’s natural ability toassociate certain ideas.Association is not “an inseparable connexion”, but rather“a gentle force, which commonly prevails”, by means ofwhich one idea naturally introduces another (T 1.1.4.1/10).

In the firstEnquiry, Hume says that even though it isobvious to everyone that our ideas are connected in this way, he isthe first philosopher who has “attempted to enumerate or classall the principles of association” (EHU 3.2/24). He regards hisuse of these “universal principles” as so distinctive thathe advertises them as his most original contribution—one thatentitles him call himself an “inventor” (Abstract35).

Hume identifies three principles of association:resemblance,contiguity in time and place, and causation. When someoneshows you a picture of your best friend, you naturally think of herbecause the pictureresembles her. When you’re remindedof something that happened in the 1960s—miniskirts, forexample—you may think of the Vietnam War, because they aretemporally contiguous. Thinking of Sausalito may lead you tothink of the Golden Gate Bridge, which may lead you to think of SanFrancisco, since they arespatially contiguous.Causality works both from cause to effect and effect tocause: meeting someone’s father may make you think of his son;encountering the son may lead you to thoughts of his father.

Of the three associative principles,causation is thestrongest, and the only one that takes us “beyond oursenses” (T 1.3.2.3/74). It establishes links between our presentand past experiences and our expectations about the future, so that“all reasonings concerning matters of fact seem to be founded onthe relation ofCause andEffect” (EHU4.1.4/26). Taking aspirin in the past has relieved my headaches, so Iexpect that the aspirin I just took will soon relieve my presentheadache. Hume also makes clear that causation is the least understoodof the associative principles, but he tells us, “we shall haveoccasion afterwards to examine it to the bottom” (T1.1.4.2/11).

Like gravitational attraction, the associative principles areoriginal, and so can’t be explained further. Althoughthe associative principles’ “effects are everywhereconspicuous” their causes “are mostly unknown, and must beresolv’d intooriginal qualities of human nature, whichI pretend not to explain”. Accordingly, we should curb any“intemperate desire” to account further for them, fordoing so would take us illegitimately beyond the bounds of experience(T 1.1.4.6/12–13).

Hume doesn’t try to explainwhy we associate ideas aswe do. He is interested only in establishing that, as a matter offact, wedo associate ideas in these ways. Given that hisclaim that the associative principles explain the important operationsof the mind is an empirical one, he must admit, as he does in thefirstEnquiry, that he cannot prove conclusively that hislist of associative principles is complete. Perhaps he has overlookedsome additional principle. We are free to examine our own thoughts todetermine whether resemblance, contiguity, and causation successfullyexplain them. The more instances the associative principles explain,the more assurance we have that Hume has identified the basicprinciples by which our minds work.

In theAbstract, Hume concludes that it should be “easyto conceive of what vast consequences these principles must be in thescience of human nature”. Since they “are the only ties ofour thoughts, they are reallyto us the cement of theuniverse, and all the operations of the mind must, in great measure,depend on them” (Abstract 35). Just what these “vastconsequences” are will become clear when we examine Hume’srevolutionary accounts of our causal inferences and moraljudgments.

5. Causation

The medieval synthesis Thomas Aquinas (1224–74) forged betweenChristian theology and Aristotle’s science and metaphysics setthe terms for the early modern causation debate. Aristotle(384–322 BCE) drew an absolute categorical distinction betweenscientific knowledge (scientia) and belief (opinio).Scientific knowledge was knowledge of causes and scientificexplanation consisted indemonstration—proving thenecessary connection between a cause and its effect fromintuitively obvious premises independently of experience.

Modern philosophers thought of themselves as scientificrevolutionaries because they rejected Aristotle’s account ofcausation. Even so, they accepted his distinction between knowledgeand belief, and regarded causal inference as an exercise of reason,which aimed at demonstrating the necessary connection between causeand effect. Malebranche (1638–1715), and others followingDescartes (1596–1650), were optimistic about the possibility ofdemonstrative scientific knowledge, while those in the Britishexperimental tradition were more pessimistic. Locke was sufficientlysceptical about what knowledge we can attain that he constructed oneof the first accounts of probable inference to show that belief canmeet standards of rationality that make experimental naturalphilosophy intellectually respectable.

When Hume enters the debate, he translates the traditional distinctionbetween knowledge and belief into his own terms, dividing “allthe objects of human reason or enquiry” into two exclusive andexhaustive categories:relations of ideas andmatters offact.

Propositions concerningrelations of ideas are intuitively ordemonstratively certain. They are knownapriori—discoverable independently of experience by“the mere operation of thought”, so their truthdoesn’t depend on anything actually existing (EHU 4.1.1/25).That the interior angles of a Euclidean triangle sum to 180degrees is true whether or not there are any Euclidean trianglesto be found in nature. Denying that proposition is a contradiction,just as it is contradictory to say that 8×7=57.

In sharp contrast, the truth of propositions concerningmatters offact depends on the way the world is. Their contraries are alwayspossible, their denials never imply contradictions, and theycan’t be established by demonstration. Asserting thatMiamiis north of Boston is false, but not contradictory. We canunderstand what someone who asserts this is saying, even if we arepuzzled about how he could have the facts so wrong.

The distinction betweenrelations of ideas andmatters offact is often called “Hume’s Fork”, generallywith the negative implication that Hume may be illicitly ruling outmeaningful propositions that don’t fit into these two categoriesor fit into both of them. To defuse this objection, however, it isimportant to bear in mind that Hume’s categories are histranslations of a traditional absolute categorical classificatorydistinction, which all his contemporaries and immediate predecessorsaccepted.

Hume’s method dictates his strategy in the causation debate. Inthecritical phase, he argues that his predecessors werewrong: our causal inferences aren’t determined by “reasonor any other operation of the understanding” (EHU 5.1.2/41). Intheconstructive phase, he supplies an alternative: theassociative principles are their basis.

Hume’s contributions to thecritical phase of thecausation debate are contained inTreatise 1.3.6 and Section4 of the firstEnquiry, appropriately titled “Scepticaldoubts concerning the operations of the understanding”. Theconstructive phase in hisEnquiry account is thefollowing section, also appropriately titled “Sceptical solutionof these doubts”, while the corresponding sections of theTreatise stretch from 1.3.7 through 1.3.10.

5.1 Causal Inference: Critical Phase

Causal inferences are the only way we can go beyond the evidence ofour senses and memories. In making them, we suppose there is someconnection between present facts and what we infer from them.But what is this connection? How is it established?

If the connection is established by an operation of reason or theunderstanding, it must concern eitherrelations of ideas ormatters of fact.

Hume argues that the connection can’t involve relations ofideas. Effects are different events from their causes, so there is nocontradiction in conceiving of a cause occurring, and its usual effectnot occurring. Ordinary causal judgments are so familiar that we tendto overlook this; they seem immediate and intuitive. But suppose youwere suddenly brought into the world as an adult, armed with theintellectual firepower of an Einstein. Could you, simply by examiningan aspirin tablet, determine that it will relieve your headache?

When we reasona priori, we consider the idea of the objectwe regard as a cause independently of any observations we have made ofit. It can’t include the idea of any otherdistinctobject, including the object we take to be its usual effect. But thenit can’t show us any “inseparable and inviolableconnection”—anynecessaryconnection—between those ideas. Trying to reasonapriori from your idea of an aspirin, without including anyinformation you have of its effects from your previous experience,yields only your simple ideas of its “sensiblequalities”—its size, shape, weight, color, smell, andtaste. It gives you no idea of what “secret powers” itmight have to produce its usual effects. Hume concludes thatapriori reasoning can’t be the source of the connectionbetween our ideas of a cause and its effect. Contrary to what themajority of his contemporaries and immediate predecessors thought,causal inferences do not concern relations of ideas.

Hume now moves to the only remaining possibility. If causal inferencesdon’t involvea priori reasoning about relations ofideas, they must concern matters of fact and experience. Whenwe’ve had many experiences of one kind of event constantlyconjoined with another, we begin to think of them as cause and effectand infer the one from the other. But even after we’ve had manyexperiences of a cause conjoined with its effect, our inferencesaren’t determined by reason or any other operation of theunderstanding.

In the past, taking aspirin has relieved my headaches, so I believethat taking aspirin will relieve the headache I’m having now.But my inference is based on the aspirin’s superficial sensiblequalities, which have nothing to do with headache relief. Even if Iassume that the aspirin has “secret powers” that are doingthe heavy lifting in relieving my headache, they can’t be thebasis of my inference, since these “secret powers” areunknown.

Nonetheless, Hume observes, “we always presume, when we see likesensible qualities, that they have like secret powers, and expect thateffects, similar to those we have experienced, will follow fromthem” (EHU 4.2.16/33). Since we neither intuit nor inferapriori that similar objects have similar secret powers, ourpresumption must be based in some way on our experience.

But our past experience only gives us information about objects asthey were when we experienced them, and our present experience onlytells us about objects we are experiencing now. Causal inferences,however, do not justrecord our past and present experiences.Theyextend orproject what we have gathered fromexperience to other objects in the future. Since it is not necessarilytrue that an object with the same sensible qualities will have thesame secret powers that past objects with those sensible qualitieshad, how do we project those experiences into the future, to otherobjects that may only appear similar to those we’ve previouslyexperienced?

Hume thinks we can get a handle on this question by considering twoclearly different propositions:

(1)

I’ve foundthat headache relief has always followed my takingaspirin;

and

(2)

Taking aspirinsimilar to the ones I’ve taken in the past will relieve mypresent headache.

There is no question that “the one proposition may be justlyinferred from the other”, and that “it is alwaysinferred”. But since their connection obviously isn’tintuitive, Hume challenges us to produce the “chain ofreasoning” that takes us from propositions like (1) topropositions like (2) (EHU 4.2.16/34).

(1) summarizes my past experience, while (2) predicts what will happenin the immediate future. The chain of reasoning I need must show mehow my past experience is relevant to my future experience. I needsome further proposition or propositions that will establish anappropriatelink orconnection between past andfuture, and take me from (1) to (2) using eitherdemonstrativereasoning, concerning relations of ideas, orprobablereasoning, concerning matters of fact.

Hume thinks it is evident that demonstrative reasoning can’tbridge the gap between (1) and (2). However unlikely it may be, we canalways intelligibly conceive of a change in the course of nature. Eventhough aspirin relieved my previous headaches, there’s nocontradiction in supposing that it won’t relieve the oneI’m having now, so the supposition of a change in the course ofnature can’t be proven false by any reasoning concerningrelations of ideas.

That leaves probable reasoning. Hume argues that there is no probablereasoning that can provide a just inference from past to future. Anyattempt to infer (2) from (1) by a probable inference will beviciously circular—it will involve supposing what we are tryingto prove.

Hume spells out the circularity this way. Any reasoning that takes usfrom (1) to (2) must employ someconnecting principle thatconnects the past with the future. Since one thing that keeps us frommoving directly from past to future is the possibility that the courseof nature might change, it seems plausible to think that theconnecting principle we need will be one that will assure us thatnature isuniform—that the course of nature won’tchange—something like thisuniformity principle:

[UP]

The future will be like the past.

Adopting [UP] will indeed allow us to go from (1) to (2). But beforewe can use it to establish that our causal inferences are determinedby reason, we need to determine our basis for adopting it. [UP] isclearly notintuitive, nor is itdemonstrable, asHume has already pointed out, so onlyprobable argumentscould establish it. But to attempt to establish [UP] this way would beto try to establish probable arguments using probable arguments, whichwill eventually include [UP] itself.

At this point, Hume has exhausted the ways reason might establish aconnection between cause and effect. He assures us that he offers his“sceptical doubts” not as a “discouragement, butrather an incitement … to attempt something more full andsatisfactory”. Having cleared the way for his constructiveaccount, Hume is ready to do just that.

5.2 Causal Inference: Constructive Phase

Hume calls his constructive account of causal inference a“sceptical solution” to the “sceptical doubts”he raised in the critical phase of his argument.

Since we’re determined—caused—to makecausal inferences, then if they aren’t “determin’dby reason”, there must be “some principle of equal weightand authority” that leads us to make them. Hume maintains thatthis principle iscustom orhabit:

whenever the repetition of any particular act or operation produces apropensity to renew the same act or operation … we always say,that this propensity is the effect ofCustom. (EHU5.1.5/43)

It is therefore custom, not reason, which “determines the mind… to suppose the future conformable to the past”(Abstract 16). But even though we have located the principle, it isimportant to see that this isn’t anew principle bywhich our minds operate.Custom andhabit aregeneral names for the principles of association.

Hume describes their operation as a causal process: custom or habit isthe cause of the particular propensity you form after your repeatedexperiences of the constant conjunction of smoke and fire. Causationis the operative associative principle here, since it is the only oneof those principles that can take us beyond our senses andmemories.

Hume concludes that custom alone “makes us expect for thefuture, a similar train of events with those which have appeared inthe past” (EHU 5.1.6/44). Custom thus turns out to be the sourceof theUniformity Principle—the belief that the futurewill be like the past.

5.3 Belief

Causal inference leads us not only toconceive of the effect,but also toexpect it. When I expect that aspirin willrelieve my headache, I’m not just abstractly considering theidea of headache relief, Ibelieve that aspirin will relieveit. What more is involved inbelieving that aspirin willrelieve my headache than in merelyconceiving that itwill?

It can’t be that beliefs have some additional idea—theidea of belief, perhaps—that conceptions lack. If there weresome such idea, given our ability to freely combine ideas, we could,by simply willing, add that idea to any conception whatsoever, andbelieve anything we like.

Hume concludes that belief must be some sentiment or feeling arousedin us independently of our wills, which accompanies those ideas thatconstitute them. It is a particular way or manner of conceiving anidea that is generated by the circumstances in which we findourselves.

If constant conjunctions were all that is involved, my thoughts aboutaspirin and headaches would only be hypothetical. For belief, one ofthe conjoined objects must be present to my senses or memories; I mustbe taking, or just have taken, an aspirin. In these circumstances,believing that my headache will soon be relieved is as unavoidable asfeeling affection for a close friend, or anger when someone harms us.“All these operations are species of natural instincts, which noreasoning … is able either to produce or prevent” (EHU5.1.8/46–47).

While Hume thinks thatdefining this sentiment may beimpossible, we candescribe belief, if only by analogy,although he was never completely satisfied with his attempts to do so.Belief is a livelier, firmer, more vivid, steady, and intenseconception of an object. Hume intends these characterizations to gobeyond merely recording intensity of feeling to capture how belief

renders realities … more present to us than fictions, causesthem to weigh more in the thought, and gives them a superior influenceon the passions and imagination. (EHU 5.2.12/49)

But how does an idea come to be conceived in such a manner that itconstitutes a belief?

Hume’s explanation is that as I become accustomed toaspirin’s relieving my headaches, I develop a propensity—atendency—to expect headache relief to follow taking aspirin. Thepropensity is due to the associative bond that my repeated experiencesof taking aspirin and headache relief have formed. My presentimpressions of taking an aspirin are as forceful and vivid as anythingcould be, and some of their force and vivacity transfers across theassociative path to the idea of headache relief, enlivening it withenough force and vivacity to give it the “strength andsolidity” that constitutes belief.

Since I don’t know how aspirin relieves headaches, it isfortunate that there is “a kind of pre-established harmonybetween the course of nature and the succession of our ideas”that teaches me to take aspirin when I have a headache. Custom, Humemaintains, in language that anticipates and influenced Darwin,

is that principle by which this correspondence has been effected; sonecessary to the subsistence of our species, and the regulation of ourconduct, in every circumstance of human life. (EHU 5.2.21/55)

It is far better, Hume concludes, to rely on “the ordinarywisdom of nature”, which ensures that we form beliefs “bysome instinct or mechanical tendency”, rather than trusting itto “the fallacious deductions of our reason” (EHU5.2.22/55).

In keeping with his project of providing a naturalistic account of howour minds work, Hume has given empirical explanations of ourpropensity to make causal inferences, and the way those inferenceslead to belief.

6. The Idea of Necessary Connection

The early modern causation debate revolved around a family of“nearly synonymous” key ideas, the most prominent of whichwere the ideas ofpower andnecessary connection.For Hume, “there are no ideas, which occur in metaphysics, moreobscure and uncertain”. He showcases the critical andconstructive uses of his account of definition as he attempts“to fix … the precise meaning of these terms”, inorder to “remove some part of that obscurity, which is so muchcomplained of in this species of philosophy” (EHU7.1.3/61–62).

6.1 Necessary Connection: Critical Phase

To get clear about the idea of power or necessary connection, we needto determine the impressions that are its source. Hume identifiesthree possible sources in the work of his predecessors: Locke thoughtwe get our idea of power secondarily fromexternalimpressions of the interactions of physical objects, andprimarily frominternal impressions of our ability to moveour bodies and to consider ideas. Malebranche argued that what we taketo be causes of the motion of bodies or mental activity aren’tcauses at all. They are onlyoccasions for God, the solesource of necessary connection, to act in the world. Hume rejects allthree possibilities.

He argues thatexternal impressions of the interactions ofbodies can’t give rise to our idea of power. When we see thatthe motion of one billiard ball follows another, we’re onlyobserving theirconjunction, never theirconnection.

Attending tointernal impressions of the operations of ourminds doesn’t help. Although voluntary bodily movements followour willing that those movements occur, this is a matter of fact Ilearn through experience, not from some internal impression of mywill’s power. When I decide to type, my fingers move over thekeyboard. When I decide to stop, they stop, but I have no idea howthis happens. Were I aware of the power of my will to move my fingers,I’d know both how it worked and its limits.

Our ability to control our thoughts doesn’t give us animpression of power, either. We don’t have a clue about how wecall up our ideas. Our command over them is limited and varies fromtime to time. We learn about these limitations and variations onlythrough experience, but the mechanisms by which they operate areunknown and incomprehensible to us. If I decide to think aboutIstanbul, my idea of that city comes to mind, but I experience onlythe succession of my decision followed by the idea’s appearance,never the power itself.

When ordinary people can’t determine an event’s cause,they attribute it to some “invisible intelligentprinciple”. Malebranche and other occasionalists do the same,except they apply it across the board. True causes aren’t powersin the physical world or in human minds. The only true cause isGod’s willing that certain objects should always be conjoinedwith certain others.

Anyone aware of our minds’ narrow limits should realize thatMalebranche’s theory takes us into“fairyland”—it goes so far beyond our experiencethat we have no way of intelligibly assessing it. It also capitalizeson how little we know about the interactions of bodies, but since ouridea of God is based on extrapolations from our faculties, ourignorance should also apply to him.

6.2 Necessary Connection: Constructive Phase

Since we’ve canvassed the leading contenders for the source ofour idea of necessary connection and found them wanting, it mightseem as if we have no such idea, but that would be too hasty.In our discussion of causal inference, we saw that when we find thatone kind of event is constantly conjoined with another, we begin toexpect the one to occur when the other does. We suppose there’ssome connection between them, and don’t hesitate to call thefirst, thecause, and the second, theeffect. Wealso saw that there’s nothing different in the repetition ofconstantly conjoined cases from the exactly similar single case,except that after we’ve experienced their constantconjunction, habit determines us to expect the effect when the causeoccurs.

Hume concludes that it is just this felt determination of themind—our awareness of this customary transition from oneassociated object to another—that is the source of our idea ofnecessary connection. When we say that one object is necessarilyconnected with another, we really mean that the objects have acquiredan associative connection in our thought that gives rise to thisinference.

Having located the missing ingredient, Hume is ready to offer adefinition ofcause. In fact, he gives us two. The first,

A cause is an object, followed by another, where all the objectssimilar to the first are followed by objects similar to thesecond,

gives the relevantexternal impressions, while thesecond,

A cause is an object followed by another, and whose appearance alwaysconveys the thought to the other,

captures theinternal impression—our awareness of beingdetermined by custom to move from cause to effect. Both aredefinitions on Hume’s account, but his “justdefinition” of our idea of cause is the conjunction of the two(EHU 7.2.29/76–77). Only together do they captureallthe relevant impressions involved.

Hume locates the source of the idea of necessary connectioninus, not in the objects themselves or even in our ideas of thoseobjects we regard as causes and effects. In doing so, he completelychanges the course of the causation debate, reversing what everyoneelse thought about the idea of necessary connection. Subsequentdiscussions of causation must confront the challenges Hume poses fortraditional, more metaphysical, ways of looking at our idea ofcausation.

Hume’s treatment of our idea of causation is his flagshipillustration of how his method works and the revolutionary results itcan achieve. He goes on to apply both his method, and its concreteresults, to other prominent debates in the modern period, includingprobable inference, testimony for miracles, free will, and intelligentdesign.

7. Moral Philosophy

Hume’s explanation of morality is an important part of hisefforts to reform philosophy. He takes his primary task to be aninvestigation into the origin of the basic moral ideas, which heassumes are the ideas of moral goodness and badness. As with the ideaof cause and necessary connection, he wants to explain moral ideas aseconomically as possible in terms of their “simplest and fewestcauses”. Determining their causes will determine what theircontent is—what we mean by them. His secondary concern is toestablish what character traits and motives are morally good andbad.

Hume follows his sentimentalist predecessor, Francis Hutcheson(1694–1746), in building his moral theory around the idea of aspectator who approves or disapproves of people’s charactertraits and motives. The sentiments of approval and disapproval are thesource of our moral ideas of goodness and badness. To evaluate acharacter trait as morally good is to evaluate it as virtuous; toevaluate it as morally bad is to evaluate it as vicious.

As he did in the causation debate, Hume steps into an ongoing debateabout ethics, often called the British Moralists debate, which beganin the mid-seventeenth century and continued until the end of theeighteenth. He uses the same method here as he did in the causationdebate: there is acritical phase in which he argues againsthis opponents, and aconstructive phase in which he developshis version of sentimentalism. Hume has two sets of opponents: theself-love theorists and the moral rationalists. He became the mostfamous proponent of sentimentalism.

Thomas Hobbes’ (1588–1679) radical attempt to derive moraland political obligation from motives of self-interest initiated theBritish Moralists debate. Hobbes, as his contemporaries understoodhim, characterizes us as naturally self-centered and power-hungry,concerned above all with our own preservation. In the state of nature,a pre-moral and pre-legal condition, we seek to preserve ourselves bytrying to dominate others. Since we are all sufficiently“equal” in power, this results in a state of “war ofall against all” in which life is “nasty, brutish, andshort” (Leviathan, Ch. 13). The way out is to make acompact with one another. We agree to hand over our power and freedomto a sovereign, who makes the laws necessary for us to live togetherpeacefully and has the power to enforce them. While acting morallyrequires that we comply with the laws the sovereign establishes, thebasis of morality is self-interest.

Bernard Mandeville’s (1670–1733)The Fable of theBees served to reinforce this reading of Hobbes during the early18^th century. According to Mandeville, human beings arenaturally selfish, headstrong, and unruly. Some clever politicians,recognizing that we would be better off living together in a civilizedsociety, took up the task of domesticating us. Realizing that we areproud creatures, highly susceptible to flattery, they were able todupe many of us to live up to the ideal of virtue—conquering ourselfish passions and helping others—by dispensing praise andblame. Moral concepts are just tools clever politicians used to tameus.

Two kinds of moral theories developed in reaction first to Hobbes andthen to Mandeville—rationalism and sentimentalism. Therationalists oppose Hobbes’ claim that there is no right orwrong in the state of nature, that rightness or wrongness isdetermined by the sovereign’s will, and that morality requiressanctions to motivate us. The sentimentalists object to Hobbes’and Mandeville’s “selfish” conceptions of humannature and morality. By the mid–eighteenth century, rationalistsand sentimentalists were arguing not only against Hobbes andMandeville, but also with each other.

Hume opposes both selfish and rationalist accounts of morality, but hecriticizes them in different works. In theTreatise, Humeassumes that Hobbes’ theory is no longer a viable option, sothat there are only two possibilities to consider. Either moralconcepts spring from reason, in which case rationalism is correct, orfrom sentiment, in which case sentimentalism is correct. If one falls,the other stands. In the secondEnquiry, Hume continues tooppose moral rationalism, but his arguments against themappear in an appendix. More importantly, he drops the assumption hemade in theTreatise and takes the selfish theories of Hobbesand Mandeville as his primary target. Once again, he thinks there areonly two possibilities. Either our approval is based in self-interestor it has a disinterested basis. The refutation of one is proof of theother.

7.1 Moral Rationalism: Critical Phase in theTreatise

Hume thinks that “systems and hypotheses” have also“perverted our natural understanding” of morality. Theviews of the moral rationalists—Samuel Clarke (1675–1729),Locke and William Wollaston (1660–1724)—are prominentamong them. One distinctive, but unhealthy, aspect of modern moralphilosophy, Hume believes, is that it allies itself with religion andthus sees itself as serving the interests of “popularsuperstition”. Clarke’s theory and those of the otherrationalists epitomize this tendency.

Clarke, Hume’s central rationalist opponent, appeals to reasonto explain almost every aspect of morality. He believes that there aredemonstrable moral relations of fitness and unfitness that we discovera priori by means of reason alone. Gratitude, for example, isa fitting or suitable response to kindness, while ingratitude is anunfitting or unsuitable response. He believes that the rationalintuition that an action is fitting has the power both to obligate usand to move us. To act morally is to act rationally.

Hume’s most famous and most important objection to moralrationalism is two-pronged. InTreatise 2.3.3, “Of theinfluencing motives of the will”, he rejects the rationalistideal of the good person as someone whose passions and actions aregoverned by reason. In T 3.1.1, he uses these arguments to show thatmoral ideas do not spring from reason alone.

In the first prong of his objection, Hume begins by remarking thatnothing is more common than for philosophers, as well as ordinarypeople, to talk about the “combat” between reason andpassion. They say we ought to be governed by reason rather thanpassion, and if our passions are not in line with reason’scommands, we ought to restrain them or bring them into conformity withreason. Hume counters that “reason alone can never be a motiveto any action of the will” and that by itself it can neveroppose a passion in the direction of the will.

His first argument rests on his empiricist conception of reason. As wesaw in his account of causation, demonstrative reasoning consists incomparing ideas to find relations among them, while probable reasoningconcerns matters of fact. He considers mathematical reasoning from therelation of ideas category and causal reasoning from the category ofmatters of fact. He asks us to look at instances of actions wherethese two types of reasoning are relevant and says that when we do, wewill see that reason alone couldn’t have moved us.

No one thinks that mathematical reasoning by itself is capable ofmoving us. Suppose you want to stay out of debt. This may move you tocalculate how much money comes in and how much goes out, butmathematical reasoning by itself does not move us to do anything.Mathematical reasoning, when it bears on action, is always used inconnection with achieving some purpose and thus in connection withcausal reasoning.

Hume, however, argues that when causal reasoning figures in theproduction of action, it always presupposes an existing desire orwant. On his view, reasoning is a process that moves you from one ideato another. If reasoning is to have motivational force, one of theideas must be tied to some desire or affection. As he says,

It can never in the least concern us to know, that such objects arecauses, and such others effects, if both the causes and effects areindifferent to us. Where the objects themselves do not affect us,their connexion can never give them any influence; and ‘tisplain, that as reason is nothing but the discovery of this connexion,it cannot be by its means that the objects are able to affect us (T2.3.3.3/414).

Noticing a causal connection between exercise and losing weight willnot move you to exercise, unless you want to lose weight.

It immediately follows that reason alone cannot oppose a passion inthe direction of the will. To oppose a passion, reason must be able togive rise to a motive by itself, since only a motive can opposeanother motive, but he has just shown that reason by itself is unableto do this.

Having exposed reason’s pretensions to rule, Hume inverts therationalist’s ideal of the good person, and concludes that“Reason is, and ought only to be the slave of the passions, andcan never pretend to any other office than to serve and obeythem” (T 2.3.3.4/415).

The second prong of Hume’s objection, the argument frommotivation, is directed primarily against Clarke and concerns thesource of our moral concepts: either they spring from reason or fromsentiment. Couching this debate in terms of his own version of thetheory of ideas, he reminds us that to engage in any sort of mentalactivity is to have a perception before the mind, so “to approveof one character, to condemn another, are only so many differentperceptions” (T 3.1.1.2/456). Since there are only two types ofperception—ideas and impressions—the question betweenrationalism and sentimentalism is

Whether ’tis by means of our ideas or impressions wedistinguish betwixt vice and virtue, and pronounce an action blameableor praise-worthy? (T 3.1.1.3/456)

The argument from motivation has only two premises. The first is thatmoral ideas have pervasive practical effects. Experience shows that weare often motivated to perform an action because we think it isobligatory or to refrain because we think it is unjust. We try tocultivate the virtues in ourselves and are proud when we succeed andashamed when we fail. If morality did not have these effects on ourpassions and actions, moral rules and precepts would be pointless, aswould our efforts to be virtuous. Thus “morals excite passions,and produce or prevent actions” (T 3.1.1.6/457).

The second premise is that by itself reason is incapable of excitingpassions or producing and preventing actions, which Hume supports withthe arguments we just looked at about the influencing motives of thewill. The argument from motivation, then, is that if moral conceptsare capable of exciting passions and producing or preventing actions,but reason alone is incapable of doing these things, then moralconcepts can’t spring from reason alone.

Reason for Hume is essentially passive and inert: it is incapable byitself of giving rise to new motives or new ideas. Although he thinksthe argument from motivation is decisive, in T 3.1.1 he offers abattery of additional arguments, which are intended to show that moralconcepts do not arise from reason alone.

Hume takes the defeat of rationalism to entail that moral conceptsspring from sentiment. Of course, he was not the first to claim thatmoral ideas arise from sentiment. Hutcheson claimed that we possess,in addition to our external senses, a special moral sense thatdisposes us to respond to benevolence with the distinctive feelings ofapprobation. Hume, however, rejects the idea that the moral sentimentsarise from a sense that is an “original quality”and part of our “primary constitution”.

He first argues that there are many different types of virtue, not allof which are types of benevolence—respecting people’sproperty rights, keeping promises, courageousness, andindustriousness—as Hutcheson maintained. If we agree with Hume,but keep Hutcheson’s idea of a moral sense, we would have tobelieve that we have many different “original” senses,which dispose us to approve of the variety of different virtuesseparately. But he complains that this is not only highly implausible,but also contrary to the

usual maxims, by which nature is conducted, where a few principlesproduce all the variety we observe in the universe. (T3.1.2.6/473)

Instead of multiplying senses, we should look for a few generalprinciples to explain our approval of the different virtues.

The real problem, however, is that Hutcheson justclaims—hypothesizes—that we possess a unique, originalmoral sense. If asked why we have a moral sense, his reply is that Godimplanted it in us. Although in his critical phase Hume freely borrowsmany of Hutcheson’s arguments to criticize moral rationalism,his rejection of a God-given moral sense puts him on a radicallydifferent path from Hutcheson in his constructive phase. One way ofunderstanding Hume’s project is to see it as an attempt tonaturalize Hutcheson’s moral sense theory. He aims to provide awholly naturalistic and economical explanation of how we come toexperience the moral sentiments that also explains why we approve ofthe different virtues. In the course of explaining the moralsentiments, Hutcheson’s idea of an original moral sensedisappears from Hume’s account of morality.

7.2 Sentimentalism: Constructive Phase

InTreatise 3.3.1, Hume turns to his constructive task ofproviding a naturalistic explanation of the moral sentiments. Herefers to them as feelings of approval or disapproval, praise orblame, esteem or contempt. Approval is a kind of pleasant or agreeablefeeling; disapproval a kind of painful or disagreeable feeling. Inseveral key passages, he describes the moral sentiments as calm formsof love and hatred. When we evaluate our own character traits, prideand humility replace love and hatred.

Hume’s project is “to discover the true origin of morals,and of that love or hatred, which arises” (T 3.3.1/575) when wecontemplate our own or other people’s character traits andmotives. He traces the moral sentiments tosympathy. Sympathyis a psychological mechanism that explains how we come to feel whatothers are feeling. It is not itself a feeling or sentiment and soshould not be confused with feelings of compassion or pity. Humeappeals to sympathy to explain a wide range of phenomena: our interestin history and current affairs, our ability to enjoy literature,movies, and novels, as well as our sociability. It is central to hisexplanations of our passions, our sense of beauty, and our sense ofwhat is morally good and bad.

Sympathy is a process that moves me from my idea of what someone isfeeling to actually experiencing the feeling. There are four steps tothis process. I first arrive at the idea of what someone is feeling inany of the usual ways. I next become aware of theresemblances between us, so we are linked by that principleof association. While we resemble every human being to some extent, wealso resemble some individuals more than others—for instance,those who share our language or culture or are the same age and sex aswe are. The associative principles ofcontiguity andcausality also relate individuals who are located closely tous in time or space or who are family members or teachers. Accordingto Hume, we are able to sympathize more easily and strongly withindividuals with whom we have strong associative ties. The strongerthe associative relations, the stronger our sympathetic responses.Hume then claims—controversially—that we always have avivid awareness of ourselves. Finally, he reminds us that theprinciples of association not only relate two perceptions, but theyalso transmit force and vivacity from one perception to another.

Suppose my friend recently suffered a devastating loss and I realizeshe is feeling sad. The associative principles transmit force andvivacity from my vivid awareness of myself to my idea of myfriend’s sadness. Since for Hume the difference betweenimpressions and ideas is that impressions are more lively andvivacious than ideas, if an idea of a passion is sufficientlyenlivened, it becomes the very passion itself. I now feel sad too, butnot quite as strongly as my friend.

The way Hume uses the idea that the associative principles transmitforce and vivacity in his explanation of sympathy is parallel to theway he uses it in his explanation of causal inference. In the case ofcausal inference, if we have an impression of an effect (smoke), theassociative principles give rise not only to the idea of its cause(fire), but they also transmit some of the impression’s forceand vivacity to the idea of its cause, so that we come to believe thatfire is the cause of the smoke. A belief is an idea that is so livelythat it islike an impression, and influences us in the wayimpressions do. Similarly, my lively awareness of myself enlivens byassociation my idea of my friend’s sadness. But the result inthe case of sympathy is even stronger: when an idea of a passion issufficiently enlivened, it becomes the very passion itself.

One advantage Hume’s explanation of the moral sentiments interms of sympathy has over Hutcheson’s claim that we possess aGod-given moral sense is that it enables him to provide a unifiedtheory of the mind. He explains the moral sentiments by appealing tosympathy, which, in turn, he explains in terms of the same associativeprinciples he invoked to explain causal beliefs. Without sympathy, andthe associative principles that explain it, we would be unimaginablydifferent than we are—creatures without causal or moralideas.

Hume develops his account of moral evaluation further in response totwo objections to his claim that the moral sentiments arise fromsympathy. The first is the “sympathy is variable”objection. Sympathy enables us to enter into the feelings of anyone,even strangers, because we resemble everyone to some extent. But it isan essential feature of his account of the natural and spontaneousoperation of sympathy that our ability to respond sympathetically toothers varies with variations in the associative relations. I am ableto sympathize more easily and strongly with someone who resembles meor is related to me by contiguity or causation. The objection is thatthe moral sentiments can’t be based in sympathy because theloves and hatreds that result from the natural and spontaneousworkings of sympathy vary, but our moral approval doesn’t vary.The second objection is that “virtue in rags” still evokesour approval. Sympathy works by looking at the actual effects of aperson’s character traits, but sometimes misfortune or lack ofopportunity may prevent an individual from exercising their goodcharacter traits, yet we still admire them.

Hume argues that moral love and hatred spring from sympathy, but onlywhen we regulate our sympathetic reactions by taking up what he calls“the general point of view”. There are two regulatoryfeatures to the general point of view. The first is that we survey aperson’s character from the perspective of the person and hisusual associates—friends, family, neighbors, and co-workers. Wesympathize with the person and the people with whom that personregularly interacts and judge character traits in terms of whetherthey are good or bad for these people. Second, we regulate sympathyfurther by relying on general rules that specify the general effectsand tendencies of character traits rather than sympathizing with theiractual effects.

By putting together these two regulatory features, we arrive atHume’s idea of the general point of view, which defines aperspective from which we may survey a person’s character traitsthat we share with everyone. When we occupy the general point of view,we sympathize with the person herself and her usual associates, andcome to admire the person for traits that are normally good foreveryone. The general point of view is, for Hume, the moralperspective. We do not experience the moral sentiments unless we havealready taken up the general point of view. The moral sentiments andthe concepts to which they give rise are products of taking up thatstandpoint.

Hume offers the claim that we admire four sorts of charactertraits—those that are useful or immediately agreeable to theagent or to others—as an empirical hypothesis. While he providessupport for it in his discussion of the individual virtues, he alsouses his fourfold classification to undermine Christian conceptions ofmorality. He makes pride a virtue and humility a vice. He throws outthe “monkish” virtues—celibacy, fasting, andpenance—on the grounds that they are not pleasant or useful toanyone. He also rejects the distinction between virtues and naturaltalents, which legislators, “divines” and modern moralistsdefend by claiming that the moral virtues are voluntary, whereasnatural talents aren’t. Their goal is to reform us—or atleast our outward behavior—making us better, when understood inChristian terms. They accordingly restrict the domain of the moral toactions that proceed from character traits because they believe onlythey can be modified, shaped, and controlled by sanctions, whiletalents can’t. Hume, however, rejects the distinction along withthe dubious function these reformers assign to morality.

Hume identifies both what has value and what makes things valuablewith features of our psychology. Our first-order sentiments, passionsand affections, as well as actions expressive of them, are what havemoral value. Our second-order reflective sentiments about our own orother people’s sentiments, passions and affections are what givethem value. On his view, morality is entirely a product of humannature.

In the “Conclusion” of the secondEnquiry, Humesummarizes his explanation of morality with a definition of virtue ormerit:

every quality of mind, which isuseful oragreeableto thepersonhimself or toothers,communicates a pleasure to the spectator, engages his esteem, and isadmitted under the honourable denomination of virtue or merit. (EPM9.1.12/277)

This is a precise parallel of his two definitions of cause in thefirstEnquiry. Both sets of definitions pick out features ofevents, and both record a spectator’s response to thoseevents.

7.3 Self-Interest Theories: Critical Phase in theEnquiry

Hume’s secondEnquiry is a sustained and systematicattack on the “selfish” or “self-love”theories of Hobbes and Mandeville. He follows Hutcheson in thinkingthat they assign two distinct roles to self-interest in their accountsof morality: first, moral approval and disapproval are based in aconcern for our own interest and, second, the motive of which weultimately approve is self-interest. Although many people during thisperiod understood Hobbes’ theory through Mandeville’slens, Hume believes it is important to distinguish them. As he seesit, Mandeville’s theory is superficial and easily dismissed.Hobbes is his main opponent.

Hume’s rejection of Hobbes’ selfish account of approvaland disapproval begins in Section II and ends in Part I of the“Conclusion” of theEnquiry. Like Hutcheson, hemistakenly supposes that Hobbes was offering a rival theory ofapproval and disapproval. We approve of people’s charactertraits when they benefit us and disapprove of them when they harm us.Hume looks at each of the four types of virtue and argues that in eachcase, our approval does not spring from a concern for our ownhappiness, but rather from sympathy.

In Section II, Hume argues that one reason we approve of benevolence,humanity, and public spiritedness is that they are useful to othersand to society. In Sections III and IV, he argues that the sole groundfor approving of justice and political allegiance is that they areuseful to society. In Section V, he asks: But useful for whom? Sinceit is obvious that it has to be “for some body’sinterest”, the question is “Whose interest then?” Heassumes there are only two possibilities: approval and disapprovalspring either from sentiments that are interested or from adisinterested source.

According to Hume, Hobbes’ “deduction of morals fromself-love” begins with our realization that we cannot subsistalone. A social order provides security, peace, and mutual protection,conditions that allow us to promote our own interests better than ifwe lived alone. Our own good is thus bound up with the maintenance ofsociety. Although Hume agrees with Hobbes up to this point, he rejectshis explanation that we approve of justice, benevolence, and humanitybecause they promote our own happiness.

Hume is confident that “the voice of nature andexperience” will show that Hobbes’ theory, understood inthis way, is mistaken. Borrowing many of Hutcheson’s arguments,he points out that if approval and disapproval were based on thoughtsabout the possible advantages and disadvantages to us ofpeople’s characters and actions, we would never feel approvaland disapproval of people from “very distant ages and remotecountries”, since they cannot possibly affect us. We would neveradmire the good deeds of our enemies or rivals, since they are hurtfulto us. We would also never approve or disapprove of charactersportrayed in novels or movies, since they are not real people andcannot possibly help or harm us. We approve of character traits andactions that are useful not because they benefit us, but because wesympathize with the benefits they bestow on others or society.

Hume next examines the remaining three types of charactertraits—those that are useful to the agent (industriousness, goodjudgment), agreeable to the agent (cheerfulness) or agreeable toothers (politeness, decency). Why, for example, do we approve ofindustriousness and good judgment, character traits that are primarilyadvantageous to the possessor? In most cases they are of absolutely nobenefit to us and, in cases of rivalry, they counteract our owninterest. We approve of these character traits not because they arebeneficial to us, but because we sympathize with the benefits theyconfer on others. Hume takes this as further evidence againstHobbes’ explanation in terms of self-interest and in support ofhis sympathy-based account.

In Part I of the “Conclusion”, Hume complains thatHobbes’ self-love theory is unable to explain two importantfeatures of our moral sentiments: we tend to approve of the same sortsof character traits and we are able to morally evaluate anyone, at anytime or place. If our approval and disapproval were based on thoughtsabout our own benefits and harms, the moral sentiments would vary fromperson to person and for the same person over time. We wouldn’thave moral feelings about most people, since most people don’taffect us. The moral sentiments spring from our capacity to respondsympathetically to others.

Hume is equally adamant that any explanation of the motives thatprompt us to virtuous actions in terms of self-interest is mistaken.He opposes them in Appendix II of theEnquiry, which wasoriginally part of Section II, “Of Benevolence”. Hefollows Hutcheson in thinking that the issue is whether the variousbenevolent affections are genuine or arise from self-interest. Onceagain he distinguishes Mandeville’s from Hobbes’explanations of benevolence and takes Hobbes to be his main opponent.On Hume’s reading of Hobbes, while we approve of kindness,friendship, and other benevolent affections, any desire to benefitothers really derives from self-interest, although we may not alwaysbe conscious of its influence on those desires.

Hume offers two arguments against this selfish view. He first asks usto consider cases in which people are motivated by a genuine concernfor others, even when such concern could not possibly benefit them andmight even harm them. We grieve when a friend dies, even if the friendneeded our help and patronage. How could our grief be based inself-interest? Parents regularly sacrifice their own interests for thesake of their children. Non-human animals care about members of theirown species and us. Is their concern a “deduction” ofself-interest? He concludes that these and “a thousand otherinstances … are marks of a general benevolence in human nature,where no interest binds us” (EPM App 2.11/300).

Hume supplements this argument from experience with a highlycompressed sketch of an argument he borrows from Butler. Happinessconsists in the pleasures that arise from the satisfaction of ourparticular appetites and desires. It is because we want food, fame,and other things that we take pleasure in getting them. If we did nothave any particular appetites or desires, we would not want anythingand there would be nothing from which we would get pleasure. To getthe pleasures that self-love aims at, we must want something otherthan happiness itself.

7.4 Justice: Constructive Phase

Hume rightly showcases his pioneering account of justice. In theTreatise, he emphasizes the distinction between the naturaland artificial virtues. The natural virtues—being humane, kind,and charitable—are character traits and patterns of behaviorthat human beings would exhibit in their natural condition, even ifthere were no social order. The artificial virtues—respectingpeople’s property rights, fidelity in keeping promises andcontracts, and allegiance to government—are dispositions basedon social practices and institutions that arise from conventions.

Hume believes that nature has supplied us with manymotives—parental love, benevolence, and generosity—thatmake it possible for us to live together peacefully in small societiesbased on kinship relations. One of his important insights is thatnature has not provided us with all the motives we need to livetogether peacefully in large societies. After arguing inTreatise 3.2.1 that justice is artificial, in T 3.2.2, heasks two different questions: What motivates human beings to establishthe rules of justice that give rise to property rights, and why do weapprove of people who obey these rules of justice? The first questionconcerns justice as a practice constituted by its rules. The secondconcerns justice as a virtue, a person’s disposition to obey therules of justice.

Hume argues that we enter into a series of conventions to bring aboutpractices, each of which is a solution to a problem. Each conventiongives rise to new problems that in turn pressure us to enter intofurther conventions. The convention to bring about property rights isonly the first of several into which we enter. After property rightsare established, we enter into conventions to transfer property and tomake promises and contracts. According to him, we are by naturecooperators, although at first we cooperate only with members of ourown family. But it is also advantageous for us to cooperate withstrangers, since it allows us to produce more goods and to exchangethem. All three conventions are prior to the formation of government.On Hume’s view, it is possible for there to be a peacefulsociety of property owners who transfer and exchange materialpossessions before there is government.

Hume argues that the practice of justice is a solution to a problem wenaturally face. The problem is that since we care most about ourfamily and close friends, but material goods are scarce and portable,we are tempted to take goods from strangers to give to our family andfriends. Disputes over these goods are inevitable, but if we quarrelwe will forfeit the benefits that result from living together insociety—increased power, ability, and security. The solution tothe problem is to establish property rights. We make rules thatspecify who has a right to what, and agree to follow the rules and tokeep our hands off the property of others. Hume was one of the firstto see that what is useful is the practice of justice, rather thanindividual acts of justice. Like Hobbes, he believes that it is in ourinterest to have the practice of justice in place.

As we just saw, Hume parts company with Hobbes when he answers thesecond question about why we approve of people who obey the rules ofjustice. If Hobbes’ answer in terms of self-interest isexcluded, he thinks only one possibility remains. We approve of justpeople not because they benefit us but because we sympathize with thebenefits they bestow on others and society as a whole. Hume thusexplains our approval of justice by appealing to the same principle heinvoked to explain our approval of the natural virtues. Thus

self-interest is the original motive to the establishmentof justice: but a sympathy with public interest is the source ofthe moral approbation,which attends that virtue. (T3.2.2.24/499–500)

In Part 2 of the “Conclusion” of theEnquiry,Hume raises a serious problem with his account of justice. While it isin our interest to have the practice of justice in place, it may notalways be in our interest to obey its rules in every case. This is thefree rider problem. The free rider, whom Hume calls the sensibleknave, wants to get the benefits that result from having a practice inplace without having to always follow its rules. He knows that theonly way to obtain the advantages of social cooperation is for thepractice of justice to be in place, but he also realizes that a singleact of injustice will not significantly damage the practice. Mostpeople will obey the rules of justice, so if he commits one act ofinjustice, the institution will not be in any danger of collapsing.Suppose he has the opportunity to commit an act of injustice that willbenefit him greatly. Why shouldn’t he?

Hume confesses that if the sensible knave expects an answer, he is notsure there is one that will convince him.

If his heart rebel not against such pernicious maxims, if he feel noreluctance to thoughts of villainy or baseness, he has indeed lost aconsiderable motive to virtue…. (EPM 9.2.23/283)

There is no general agreement about whether Hume actually provides ananswer to the sensible knave and if he does, whether it isadequate.

8. Philosophy of Religion

Hume wrote forcefully and incisively on almost every central questionin the philosophy of religion, contributing to ongoing debates aboutthe reliability of reports of miracles, the immateriality andimmortality of the soul, the morality of suicide, and the naturalhistory of religion, among others. All his work excited heatedreactions from his contemporaries, and his arguments still figurecentrally in discussions of these issues today.

Hume’s greatest achievement in the philosophy of religion is theDialogues concerning Natural Religion, which is generallyregarded as one of the most important and influential contributions tothis area of philosophy. While all Hume’s books provokedcontroversy, theDialogues were thought to be so inflammatorythat his friends persuaded him to withhold them from publication untilafter his death.

Hume’s philosophical project, and the method he developed toexecute it, dictates his strategy in all the debates he entered. Inthe debates about causation and ethics, there is an initialcritical phase, where Hume assesses the arguments of hispredecessors and contemporaries, followed by aconstructivephase, where he develops his own position. In the naturalreligion debate, however, the situation is very different.Hume’s critique of the central concepts of natural religion inthecritical phase shows that these concepts have no content,so there is nothing for theconstructive phase of hisargument to be about. Instead ofresolving this debate, Humeeffectivelydissolves it.

TheDialogues are a sustained and penetrating criticalexamination of a prominent argument from analogy for the existence andnature of God, the argument from design. The argument from designattempts to establish that the order we find in the universe is solike the order we find in the products of human artifice that it toomust be the product of an intelligent designer.

8.1 The Characters

TheDialogues record a conversation between three characters.Cleanthes, a self–­proclaimed “experimentaltheist”, offers the argument from design as an empirical proofof God’s existence and nature (DCNR 5.2/41).Demeaopposes him, maintaining that the argument’s merely probableconclusion demeans God’s mystery and majesty. He believes thatGod’s nature is completely inscrutable. Cleanthes dubs Demea amystic, while Demea derides Cleanthes’anthromorphism—his human–centered bias incomparing the creator of the universe to a human mind.

Cleanthes and Demea represent the central positions in theeighteenth–century natural religion debate. Cleanthes embodiesits dominant, progressive strain, consisting primarily of theologiansin the British Royal Society, who were fascinated by probability andthe previous century’s impressive successes in experimentalnatural philosophy. Convinced that the new science gave witness toGod’s providence, they rejected traditionala prioriproofs, which purported to demonstrate God’s existence withmathematical certainty and without appeal to experience. Instead, theyused the order and regularity they found in the universe to constructa probabilistic argument for a divine designer.

Holdouts clung to demonstrative proof in science and theology againstthe rising tide of probability. Demea is the champion of theseconservative traditionalists. Since he trots out a lame version ofSamuel Clarke’s cosmological argument in Part 9, some havethought that Hume models Demea on him. But Demea lacks Clarke’srigid rationalism. It is more likely that he epitomizes a group ofminor theologians such as William King, who stressed God’sincomprehensibility and resorted toa priori arguments onlywhen they absolutely needed them.

There was no genuinely sceptical presence in theeighteenth–century natural religion debate. This makesPhilo, who both Cleanthes and Demea characterize as asceptic, the ringer in the conversation. Although all threecharacters say very Humean things at one time or another,Philo’s views are consistently the closest to Hume’s.Philo’s form of scepticism is themitigated scepticismof the firstEnquiry, which makes him the most likelycandidate for Hume’s spokesman.

As theDialogues begin, all three characters agree that theirsubject is God’snature, since everyone agrees that heexists. Parts 1–8 concern God’snaturalattributes, his omnipotence, omniscience, and providence, whileParts 10 and 11 consider hismoral attributes, hisbenevolence and righteousness.

8.2 God’s Natural Attributes

Demea holds that God is completely unknown and incomprehensible; allwe can say is that God is a being without restriction, absolutelyinfinite and universal. Cleanthes is adamant that the argument fromdesign establishes all of God’s traditional attributes. Naturalobjects and human artifacts resemble one another, so by analogy, theircauses also resemble each other. God is therefore like a human mind,only very much greater in every respect.

Demea objects that the argument’s conclusion is only probable,but Philo responds that the real problem is that the analogy is soweak. He launches a battery of arguments to show just how weak it is.The dissimilarities between human artifacts and the universe are morestriking than their similarities. We only experience a tiny part ofthe universe for a short time; much of what we do experience isunknown to us. How can we legitimately infer anything about remoteparts of the universe, much less the universe as a whole?

Philo, however, moves quickly away from chipping at theargument’sstrength to questioning the intelligibilityof its conclusion. We have noexperience of the origin of auniverse. Since causal inference requires a basis in experiencedconstant conjunction between two kinds of things, how can welegitimately draw any conclusion whatsoever about the origin of theuniverse? Does it even require a cause? One or many? Does the cause ofthe universe itself require a cause? The problem, then, is not justthat the analogy is weak; the real problem is that it attempts to takeus beyond what we can know.

Meanwhile, Demea derides Cleanthes’anthropomorphismwhile remaining smugly satisfied with what Cleanthes disparaginglycalls hismysticism. The barbs they throw at each other, andthe speeches Philo goads them to make, help create a dilemma thatPhilo uses them to construct. He directs the dilemma at Cleanthes, butit affects both characters, although Demea is slow to realize this. Hethinks Philo is in league with him in detailing the problems withCleanthes’ anthropomorphism.

Challenging Cleanthes to explain what he means by God’s mind,Philo pushes him to admit that he means “a mind like thehuman”. Cleanthes, taking the bait, responds, “I know ofno other” (DCNR 5.4/42). He argues that mystics like Demea areno better thanatheists, since they make God so remote andincomprehensible that he bears no resemblance to humancharacteristics. Philo adds that although we regard God as perfect,perfection—as we understand it—is relative, not absolute,so we can’t conclude that we grasp God’s perfections.Since all God’s attributes involve perfection—perfectknowledge, perfect power, perfect goodness—we shouldn’tthink thatany of his attributes resemble or are evenanalogous to ours. But this means that we don’t know whatwe’re talking about when we talk about God using the familiarterms we apply to human minds.

Demea adds that giving God human characteristics, even if they aregreatly magnified, denies him attributes theists have always ascribedto him. How can an anthropomorphic God have theunity,simplicity, andimmutability of the God oftraditional theism?

Philo continues to detail just howinconvenientCleanthes’ anthropomorphism really is. If he accepts theargument from design, he must be committed to a God who is finite inall respects. But what does it mean to say that God isfinitelyperfect? Once you admit that God is finite, you’ve opened acan of worms, for there are all sorts of equally probable alternativesto intelligent design. Why think that the universe is more like ahuman artifact than an animal or a vegetable? To illustrate, Philothrows out a number of outlandish alternative hypotheses. Forinstance, if you were a spider on a planet of spiders, wouldn’tyou naturally believe that a giant spider spun an immense web tocreate the world?

Cleanthes’ design hypothesis is so underdetermined by theevidence that the only reasonable approach is to abandon any attemptto adjudicate among it and its many alternatives. Total suspension ofjudgment is the only reasonable response. Otherwise, we go beyond thebounds of anything to which we can give specific content.

The dilemma Philo has constructed encapsulates the issue about thecontent of the idea of God that is central to thecriticalaspect of Hume’s project in theDialogues. If youaccept that God’s attributes are infinitely perfect, you areusing ordinary terms without their ordinary meaning, so that they donot have any clear meaning. If you deny God’s infiniteperfection, you can give him understandable attributes, but onlybecause they are amplified human characteristics. The closer Cleanthescomes to regarding God’s mind as like a human mind, the closerhe comes to regarding God’s attributes as being like humanattributes, and the less Godlike his “God” is. We can onlygive the idea of God intelligible content at the perilously high costof denying that he is really God. To do so is to abandon God for somekind of superhero.

At the end of Part 8, which concludes their discussion of God’snatural attributes, Demea still thinks that Philo and he are partners.He remains clueless about Philo’s strategy until the very end ofPart 11, when he finally realizes that he too is caught in the trapPhilo has sprung.

Demea offers ana priori alternative to the design argumentin Part 9. As noted earlier, it is an abbreviated, watered–downversion of Clarke’s cosmological argument. Although Cleanthesquickly scotches his lame efforts, Part 9 serves as an interludebetween the previous discussion of God’snaturalattributes and the consideration of hismoral attributesin Parts 10 and 11.

8.3 God’s Moral Attributes

Demea begins the discussion in Part 10. Attempting to save face fromhis recent drubbing, he suggests that we don’t accept the truthsof religion as a result of reasoning, but from what wefeelwhen confronted with how helpless and miserable we are. Religion isbased on feelings of fear and anxiety that arise from awareness of our“imbecility and misery” (DCNR 10.1/68). Our forms ofworship are attempts to appease unknown powers that oppress andtorment us.

Philo joins in, claiming he is convinced that

the best and indeed the only method of bringing everyone to a duesense of religion is by just representations of the misery andwickedness of men. (DCNR 10.2/68)

They proceed with a joint litany of the misery and melancholy of thehuman condition, topping each other with catalogues of woes. Demeadoes not realize that Philo may mean very different things by“just representation” and “due sense ofreligion” than he does, so he fails to realize that Philo isjust egging him on.

Philo maintains that we can’t evade the facts of disease,famine, and pestilence, except by “apologies, which stillfarther aggravate the charge” (DCNR 10.16/72). These apologiesaretheodicies—systematic attempts to reconcileGod’s goodness with the existence of evil. Demea is alsoscornful of theodicies, blissfully unaware that all too soon he willbe offering his own.

Cleanthes finally breaks in to say thathe doesn’t feeloppressively anxious or miserable, and hopes that anguish isn’tas common as they claim. Buthoping that the extent of humanmisery is not so widespread is not the same asproving thatit is. Cleanthes is on weak ground. Philo capitalizes on it,challenging Cleanthes to explain how God’s mercy and benevolencecan possibly resemble human mercy and benevolence. Given God’somnipotence, whatever he wills happens, but neither humans nor animalsare happy, so God presumably does not will their happiness.

Cleanthes—“smiling”—grants that if Philo canprove that mankind is “unhappy or corrupted”, hewill have succeeded in doing in religion (DCNR 10.28/74). He thinks hefinally has Philo on the ropes. In forcing a sceptic to prove apositive thesis, he must not only succeed at a difficult task, butviolates his scepticism in the process. Cleanthes fails to realizethat Philo will make his case without needing to prove anything, nordoes he realize thathe will soon be the one who needs aproof.

Demea objects that Cleanthes exaggerates the dire consequences ofacknowledging the human condition, and, despite his earlier vehementrejection of theodicies, offers his own. Sometimes called “theporch view”, Demea’s theodicy compares our experience ofthe world to the world as a whole, including the afterlife, to tryingto determine the structure of a large building from what little we cansee from its porch. From our perspective, we suffer, but from a longerview, either we don’t suffer at all, or else our suffering isfor our greater good or for the greater good of the world.

Cleanthes retorts that Demea denies the facts, and offers only emptyhypotheses, which, if intelligible at all, could only establish theirbare possibility, but never their reality. The only way to respond tochallenges to God’s benevolence is to deny that the humancondition is really so miserable.

Cleanthes has now put himself in the position in which he thought hehad put Philo. He must establish that the facts are as he claims, andPhilo is quick to stress how difficult this will be. By resting hiscase on such an uncertain point, any conclusion he draws will beequally uncertain.

Philo then ups the ante by granting for the sake of argument thathuman happiness exceeds human misery. But if God is infinitelypowerful, wise, and good, why is there any misery at all? There is noanswer that preserves all God’s attributes, except to grant thatthe subject exceeds the limits of our understanding.

Philo, however, refrains from pressing the question ofintelligibility; he is more interested in building an evenstronger case against Cleanthes’ inference to God’sbenevolence. Raising the ante higher still, he grants thatpain and suffering arecompatible with God’s infinitepower and goodness. Cleanthes, however, must prove from the“mixed and confused phenomena” that God’sbenevolence isactual, not merelypossible. Doing sois doubly difficult, since any inference from finite to infinite isshaky at best, even when the data are “pure and unmixed”(DCNR 10.35/77).

Philo concludes by admitting, with less than complete sincerity, thatwhile he was hard pressed to make his case against Cleanthes when thediscussion concerned God’s natural attributes, where his moralattributes are concerned, he is at ease. He challenges Cleanthes“to tug the laboring oar” and explain how he can inferGod’s moral attributes from the facts about the human condition(DCNR 10.36/77).

Cleanthes “tugs”, but only for one short paragraph. Headmits that if we go beyond their usual meanings when we apply humanterms to God, what we say is indeed unintelligible. Abandoning allhuman analogy is thus to abandon natural religion, but preserving itmakes it impossible to reconcile evil with an infinite God.

Cleanthes realizes he has painted himself into a corner, but onceagain he thinks there is a way out. Abandon God’s infinity;think of him as “finitely perfect”. Then“benevolence, regulated by wisdom, and limited by necessity, mayproduce just such a world as the present” (DCNR 11.1/78).

Cleanthes doesn’t realize that his new theory is worse than hisold one. He also doesn’t seem to remember Philo’s earlierquestion about what “finitely perfect” might possiblymean. Instead of God, he is now committed to some kind of superhero.Besides, the story he is telling is itself a theodicy. Hissuperhero’s limitations explain why he cannot eliminate evil, orcreate an evil–free world.

In any case, Cleanthes is no better off than he was before.Conjectures may show that the data areconsistent with theidea of God, but are never sufficient to prove that he actuallyexists.

Philo then proceeds to outline four possible hypotheses about thecause of the universe: it is perfectly good; it is perfectly evil; itis both good and evil; it is neither good nor evil. Given the evil weknow exists, the data is at bestmixed, so we can’testablish either of the first two hypotheses. The regularity anduniformity of the general laws we find in experience is sufficient todiscount the third, so the fourth seems the most probable. On thathypothesis, the cause of the universe is entirely indifferent to theamount of good and evil in the world.

These points aboutnatural evil also apply tomoralevil. We have no more reason to think that God’s righteousnessresembles human righteousness than we have to think that hisbenevolence resembles human benevolence. We have even less reason, infact, since moral evil outweighs moral goodness more than natural eviloutweighs natural goodness.

In addition, Cleanthes’ new form of anthropomorphism is saddledwith tracing moral evil back to God. Since every effect must have acause, either the chain of causes goes back infinitely, or it stopswith the original principle that is the ultimate cause of allthings—God.

At this point, Demea, who has become increasingly agitated duringPhilo’s speech, interrupts. He finally realizes that the casePhilo is making cuts against his own view as much as it cuts againstCleanthes’. Although it might appear that Demea can retreat tosome form of the theodicy he sketched earlier, the extent to whichPhilo’s argument upsets him suggests that he now realizes it isinadequate. If he leans on the mystery–mongering he hasprofessed until now, Philo has shown that, because of its lack ofspecific content, it does not point exclusively to a good God. It mayjust as well commit him to a supreme being who is “beyond goodand evil” and is totally indifferent to morality. Commitmentwithout content turns out to be no commitment at all. Demea realizesthis, dimly at least, as he leaves the conversation.

8.4 The Conclusion: Part 12

With Demea’s departure, Cleanthes and Philo are left to finishthe conversation. Their tone is conciliatory, so conciliatory thatPhilo says he must “confess” that although he is lesscautious about natural religion than any other subject,

no one has a deeper sense of religion impressed on his mind, or paysmore profound adoration to the divine Being, as he discovers himselfto reason, in the inexplicable contrivance and artifice of nature.(DCNR 12.2/89)

Philo’s “confession” paves the way for a blockbusterthat has puzzled generations of readers. Philo seems to reverse field,apparently recanting what he has argued for so forcefully. He grantsCleanthes that “a purpose, an intention, a design, strikeseverywhere the most careless, the most stupid thinker” (DCNR12.2/89).

His remarks are, however, by no means straightforward. Some takePhilo—and, by implication, Hume—to be outing himself as acloset theist. Others conclude that, since he holds all the cards atthis point, he can afford to be conciliatory. Read ironically, Philocould be saying that while “careless and stupid” observersare struck by purpose, intention, and design in the universe, careful,critical, intelligent ones are not. But there is no need to force theirony here. Read straight, nature’s “contrivance andartifice” is “inexplicable” precisely because reasoncan discover nothing about God’s natural or moral attributes.Everyone—even the stupid and careless—can see that theparts of animals and plants havefunctions, and so can easilyunderstand why “an anatomist, who discovered a new organ orcanal, would never be satisfied until he had also discovered its useand intention” (DCNR 12.2/90).

Recognizing that an organism’s parts haveuses—functions—says nothing aboutwhether their uses or functions are due to a designer’s plan, soPhilo’s acknowledgement implies nothing about whether he nowaccepts the design hypothesis. In fact, what he says here reiterateshis position in Part 8, that function alone is no proof of divinedesign:

it is in vain … to insist on the uses of the parts of animalsor vegetables and their curious adjustment to each other. I would fainknow how an animal could subsist, unless its parts were so adjusted?(DCNR 8.9/61)

No one should deny designin this sense, so long as they doso “without any religious purpose” (DCNR 12.2/90). Farfrom reversing himself, then, Philo’s position is continuouswith the line he has taken throughout theDialogues.

As the conversation continues, Philo provides a diagnosis of thedispute. While the works of nature do bear “a greatanalogy” to the products of human artifice, as its proponentsclaim, there are also considerable differences. He suspects that thismay be the source of the intractability of the controversy, whichsuggests that it may be at bottom “somewhat of a dispute ofwords” (DCNR 12.6/92).

But verbal disputes can be resolved—ordissolved—by providing clear definitions. However, thedilemma about the content of our idea of God that Philo hasconstructed clearly implies that such aconstructive solutionis not possible here.

Philo explains why only acritical solution is possible byoffering a deeper diagnosis of the problem. Although the dispute mayappear to be merely verbal, it is in fact “still more incurablyambiguous”, for

there is a species of controversy, which, from the very nature oflanguage and of human ideas, is involved in perpetual ambiguity, andcan never, by any precaution or any definitions, be able to reach areasonable certainty or precision. These are the controversiesconcerning the degrees of any quality or circumstance. (DCNR12.7/92)

This is exactly what the dispute over intelligent design is about.Analogies are always matters of degree, and the degrees of thequalities involved in the design argument aren’t capable ofexact measurement. The controversy thus “admits not of anyprecise meaning, nor consequently of any determination” (DCNR12.7/93). The dispute about design is actuallyworse than averbal dispute.

That is why anyone, even an atheist, can say, with equal plausibility,that “the rotting of a turnip, the generation of an animal, andthe structure of human thought” all “probably bear someremote analogy to each other” (DCNR 12.7/93). That is why Philo,without renouncing any of his previous claims, can assent to the“somewhat ambiguous, at least undefined”, and, as we haveseen,indefinable proposition into which

the whole of natural theology … resolves itself …that the cause or causes of order in the universe probably bearsome remote analogy to human intelligence. (DCNR 12.33/101)

Anything is like anything else insome remote respect. So theordering principle of the universe, if indeed there is one, can beabsolutely anything.

If this is all there is to “the whole of naturaltheology”, then we can certainly conclude that theargument’s conclusion has no religiously significant content.But it has no religiously significant content because Philo’scritique has drained it of any content whatsoever. Cleanthes’design hypothesis is not just false; it is unintelligible.

The conversation began with all three participants agreeing that theirtopic was to discuss only God’snature, not hisexistence. As it concludes, it is no longer clear that thesequestions are really so distinct as originally assumed. We don’tknow what we’re talking about when we talk about a God whosenature is inconceivable, incomprehensible, indeterminate, andindefinable. What, then, are we to make of the claim about hisexistence?

TheDialogues draw out the consequences of Hume’sstatement, in the firstEnquiry, that

the idea of God, as meaning an infinitely intelligent, wise, and goodBeing, arises from reflecting on the operations of our own mind, andaugmenting, without limit, those qualities of goodness and wisdom.(EHU 2.6/19)

If we insist on “augmenting without limit”, we let loosethe moorings that give intelligible content to God’sintelligence, wisdom, and goodness. If we stop short of the limit, wemay have content, but we have also lost God.

Bibliography

Primary Literature

Hume’s Works

The standard critical edition of Hume’s philosophical writingsisThe Clarendon Edition of the Works of David Hume,currently in progress. The General Editors are Tom L. Beauchamp, thelate David Fate Norton, and the late M.A. Stewart. The followingvolumes, in order of publication, are now in print:

• An Enquiry concerning the Principles of Morals, editedby Tom L. Beauchamp, Oxford: Clarendon Press, 1998.
• An Enquiry concerning Human Understanding, edited by Tom. L.Beauchamp, Oxford: Clarendon Press, 1999.
• A Treatise of Human Nature, edited by David Fate Nortonand Mary J. Norton, Oxford: Clarendon Press, 2000.
• A Dissertation on the Passions and The Natural History ofReligion, edited by Tom L. Beauchamp, Oxford: Clarendon Press, 2007.
• Essays, Moral, Political, and Literary, edited by Tom L.Beauchamp and Mark A. Box, Oxford: Clarendon Press, 2021.

Oxford is also simultaneously keeping these two long-familiar editionsof theTreatise and theEnquiries in print, in orderto “ensure their continued availability”:

• [T]A Treatise of Human Nature, edited byL. A. Selby-Bigge, 2^nd ed. revised by P. H. Nidditch,Oxford: Clarendon Press, 1975. Page references in this entry are tothis edition. [It also contains Hume’s “Abstract” oftheTreatise. Abbreviated [Abstract], with paragraphreferences also to this edition.]
• [EHU]An Enquiry concerning Human Understanding, and[EPM]An Enquiry concerning the Principles of Morals. Bothare contained inEnquiries concerning Human Understanding andconcerning the Principles of Morals, edited by L. A. Selby-Bigge,3^rd edition, revised by P. H. Nidditch, Oxford: Clarendon Press,1975. Page references in this entry are to this edition.

Other Works by Hume

• The Letters of David Hume, edited by J.Y.T. Grieg, 2 volumes,Oxford: Clarendon Press, 1932. [HL]
• “My Own Life” (Hume’s Autobiographical Essay),in HL 1–7].
• New Letters of David Hume, edited by Raymond Klibansky andErnest C. Mossner, Oxford: Clarendon Press, 1954.
• Further Letters of David Hume, edited by Felix Waldman,Edinburgh: Edinburgh Bibliographical Society, 2014.
• Dialogues concerning Natural Religion, edited by DorothyColeman, Cambridge: Cambridge University Press, 2007. [Abbreviated[DCNR], wirth page references to this edition.]

Other Primary Literature

• Bentham, J., 1776,A Fragment on Government, J.H. Burnsand H.L.A. Hart (eds.), Cambridge: Cambridge University Press,1977.
• Hobbes, T., 1651,Leviathan, in E. Curley (ed.),Leviathan, with selected variants from the Latin edition of1668, Indianapolis: Hackett, 1994.
• Kant, I., 1783,Prolegomena to Any Future Metaphysics ThatWill Be Able to Come Forward as a Science, in H. Allison andP. Heath (eds.),Immanuel Kant: Theoretical Philosophy after1781, Cambridge: Cambridge University Press, 2002,19–170

Secondary Literature

• Ainslie, D.C., 2015,Hume’s True Scepticism,Oxford: Oxford University Press.
• Ainslie, D.C., and Annemarie Butler (eds.), 2015,TheCambridge Companion to Hume’s Treatise, Cambridge:Cambridge University Press.
• Allison, H.E., 2008,Custom and Reason in Hume, Oxford:Oxford University Press.
• Árdal, P.S., 1966,Passion and Value inHume’s Treatise, Edinburgh: Edinburgh University Press.2^nd edition, revised, 1989.
• Baier, A.C., 1991,A Progress of Sentiments: Reflections onHume’s Treatise, Cambridge, MA: HarvardUniversity Press.
• –––, 2008,Death and Character: FurtherReflections on Hume, Cambridge, MA: Harvard University Press.
• Baxter, D.L.M., 2008,Hume’s Difficulty, London:Routledge.
• Beauchamp, T.L. and A. Rosenberg, 1981,Hume and the Problemof Causation, New York: Oxford University Press.
• Bennett, J., 2001,Learning from Six Philosophers, TwoVolumes, Oxford: Oxford University Press.
• Blackburn, S., 1993,Essays in Quasi-Realism, New York:Oxford University Press.
• Bricke, J., 1980,Hume’s Philosophy of Mind,Princeton: Princeton University Press.
• Box, M. A., 1990,The Suasive Art of David Hume,Princeton: Princeton University Press.
• Brown, C. and W. E. Morris, 2012,Starting with Hume,London: Continuum.
• Buckle, S., 2001,Hume’s Enlightenment Tract: The Unityand Purpose of An Enquiry concerning HumanUnderstanding, Oxford: Clarendon Press.
• Cohon, R. (ed.), 2001,Hume: Moral and PoliticalPhilosophy, Aldershot: England and Burlington, Vermont:Dartmouth/Ashgate.
• –––, 2008,Hume’s Morality. NewYork/Oxford: Oxford University Press.
• Costelloe, T.M., 2018,The Imagination in Hume’sPhilosophy: The Canvas of the Mind, Edinburgh: EdinburghUniversity Press.
• De Pierris, G., 2015,Ideas, Evidence, and Method:Hume’s Skepticism and Naturalism concerning Knowledge andCausation, Oxford: Oxford University Press.
• Dicker, G., 1998,Hume’s Epistemology and Metaphysics:An Introduction, London and New York: Routledge.
• Earman, J., 2000,Hume’s Abject Failure: The ArgumentAgainst Miracles, New York: Oxford University Press.
• Fodor, J.A., 2003,Hume Variations, Oxford: ClarendonPress.
• Fogelin, R.J., 1985,Hume’s Scepticism in theTreatise of Human Nature, London: Routledge andKegan Paul.
• –––, 2003A Defense of Hume onMiracles, Princeton: Princeton University Press.
• –––, 2009,Hume’s SkepticalCrisis, Oxford: Oxford University Press.
• Frasca-Spada, M., 1998,Space and the Self in Hume’sTreatise, Cambridge: Cambridge UniversityPress.
• Frasca-Spada, M. and P.J.E. Kail (eds.), 2005,Impressions ofHume, Oxford: Clarendon Press.
• Garfield, J.L., 2019,The Concealed Influence ofCustom: Hume’s Treatisefrom the InsideOut, Oxford: Oxford University Press.
• Garrett, D., 1996,Cognition and Commitment in Hume’sPhilosophy, Oxford/New York: Oxford University Press.
• –––, 2015,Hume, London and New York:Routledge.
• Harris, J.A., 2005,Of Liberty and Necessity: The Free WillDebate in Eighteenth-Century British Philosophy, Oxford:Clarendon Press.
• –––, 2015,Hume: An IntellectualBiography, Cambridge: Cambridge University Press.
• Holden, T., 2010,Spectres of False Divinity: Hume’sMoral Atheism, Oxford: Oxford University Press.
• Jones, P., 1982,Hume’s Sentiments, Edinburgh:Edinburgh University Press.
• Kail, P.J.E., 2007,Projection and Realism in Hume’sPhilosophy, Oxford: Oxford University Press.
• Livingston, D.W., 1984,Hume’s Philosophy of CommonLife, Chicago: University of Chicago Press.
• –––, 1998,Philosophical Melancholy andDelirium: Hume’s Pathology of Philosophy, Chicago:University of Chicago Press.
• Loeb, L.E., 2002,Stability and Justification inHume’s Treatise, New York: Oxford University Press.
• Millican, P. (ed.), 2002,Reading Hume on HumanUnderstanding, Oxford: Clarendon Press.
• Mossner, E.C., 1954,The Life of David Hume, London:Nelson.
• Noonan, H.W., 1999,Hume on Knowledge, London and NewYork: Routledge.
• Norton, D.F., 1982,David Hume, Common Sense Moralist,Sceptical Metaphysician, Princeton: Princeton UniversityPress.
• Norton, D. F. and J. Taylor (eds.), 2009,The CambridgeCompanion to Hume, 2^nd edition, Cambridge: CambridgeUniversity Press.
• Noxon, J., 1973,Hume’s Philosophical Development,Oxford: Oxford University Press.
• Owen, D., 2000,Hume’s Reason, Oxford: OxfordUniversity Press.
• Passmore, J., 1952,Hume’s Intentions, Cambridge:Cambridge University Press.
• Pears, D., 1990,Hume’s System, Oxford: OxfordUniversity Press.
• Penelhum, T., 1975,Hume, London: Macmillan.
• –––, 2000,Themes in Hume: The Will, TheSelf, Religion, Oxford: Clarendon Press.
• Price, H.H., 1940,Hume’s Theory of the ExternalWorld, Oxford: Clarendon Press.
• Qu, H.M., 2020,Hume’s EpistemologicalEvolution, Oxford: Oxford University Press.
• Radcliffe, E.S. (ed.), 2008,A Companion to Hume, Oxford:Blackwell.
• –––, 2018,Hume, Passion, and Action,Oxford: Oxford University Press.
• Read, R. and K.A. Richman (ed.), 2000,The New HumeDebate, New York and London: Routledge.
• Russell, P., 1995,Freedom and Moral Sentiment, New York:Oxford University Press.
• –––, 2008,The Riddle of Hume’sTreatise:Skepticism, Naturalism, and Irreligion, Oxford: OxfordUniversity Press.
• –––, 2016,The Oxford Handbook of Hume,Oxford: Oxford University Press.
• Schabas, M. and Carl Wennerlund, 2020,A Philosopher’sEconomist: Hume and the Rise of Capitalism, Chicago: Universityof Chicago Press.
• Schmitt, F.F., 2014,Hume’s Epistemology in theTreatise:A Veritistic Interpretation, Oxford: OxfordUniversity Press.
• Smith, N.K., 1941,The Philosophy of David Hume, London:Macmillan.
• Stanistreet, P., 2002,Hume’s Scepticism and the Scienceof Human Nature, Aldershot: Ashgate.
• Stewart, M.A. and J.P. Wright (eds.), 1994,Hume andHume’s Connexions, Edinburgh: Edinburgh UniversityPress.
• Strawson, G., 1989,The Secret Connexion: Causation, Realismand David Hume, Oxford: Oxford University Press.
• Stroud, B., 1977,Hume, London: Routledge and KeganPaul.
• Taylor, J.A., 2015,Reflecting Subjects: Passion, Sympathy,and Society in Hume’s Philosophy, Oxford: Oxford UniversityPress.
• Traiger, S. (ed.), 2006,The Blackwell Guide to Hume’sTreatise, Oxford: Blackwell.
• Tweyman, S., 1995,David Hume: Critical Assessments, SixVolumes, London and New York: Routledge.
• Watkins, M., 2020,The Philosophical Progress ofHume’s Essays, Cambridge: Cambridge University Press.
• Waxman, W., 1994,Hume’s Theory ofConsciousness, Cambridge: Cambridge University Press.
• Wright, J. P., 1983,The Sceptical Realism of David Hume,Minneapolis: University of Minnesota Press.
• –––, 2009,Hume’s “A Treatise ofHuman Nature”: An Introduction, Cambridge: CambridgeUniversity Press.

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

• Hume Texts Online.
• The Hume Society

Related Entries

Berkeley, George |Clarke, Samuel |free rider problem |Hobbes, Thomas |Hume, David: aesthetics |Hume, David: moral philosophy |Hume, David: Newtonianism and Anti-Newtonianism |Hume, David: on free will |Hume, David: on religion |induction: problem of |Kant, Immanuel: and Hume on causality |Kant, Immanuel: and Hume on morality |Locke, John |miracles |Scottish Philosophy: in the 18th Century

Acknowledgments

Thanks to the late Annette Baier, and to Arthur Morton and David Owen,for their assistance. The editors thank Sally Ferguson for notifyingus of a number of typographical errors.