Dr. Ari Santas’ Notes on

Hume’s Enquiry, IV: Skeptical Doubts

Concerning the Operations of the Understanding


Part I


A.     Goal of the Section

·  Recall that Hume, as an epistemologist, wants to be able to evaluate ideas and influences

      ·  In section II, he described the principles by which we relate (associate) ideas

            ·  These relations are the foundations of inference

·  In this section, he is going to focus on these principles - causation - and try to determine how we make causal inferences

      ·  By a series of questions, he will try to establish what causation is not foundation

            ·  It is not foundation on the Principles of Reason - not a priori

            ·  Contra the Rationalist Tradition

      ·  Later, in section V, he will tell you what it is founded on

            ·  Experience. . .but there is more to it


B.     A Distinction

      ·  Hume begins with a common distinction used by epsitemologists in his time

            ·  All human inquiry can be divided into two main types:

1)      Relations of Ideas (Leibniz’ Truth of Reason)

2)      Matters of Fact (Leibniz’ Truth of Fact)

·  Relation of Ideas concerns knowledge given a priori independent of (unassailable by) sense experience

      ·  Geometry, Algebra, Arithmetic are branches of this

      ·  Truths are demonstrated by logical inference and hold necessarily

      ·  Denying such truths lead to contradiction

·  Matters of Fact concerns knowledge given a posteriori - by virtue of experience

      ·  All inexact sciences fall here

·  Truths are not demonstrated but verified by appeal to experience and observation

·  Denying such truths does not lead to contradiction


C.     Focus on Matters of Fact

·  Since Hume is ultimately concerned with things concerning the word, he will consider the are of human inquiry which deals with matters of fact

·  The first question Hume asks is: what is the basis for all our reasonings concerning matters of fact?

      ·  By virtue of what do we make our inferences when reasoning about the world?

·  What can we use as evidence for any claim about the existence of something or fact beyond direct sensory awareness or memory?

·  When I tell you that my car is in the lot, what can I offer as evidence short of going out to look?

·  How is it that I can tell you that it will rain this weekend, or that the sun will rise tomorrow?


D.     Causality

·  The answer is that all our reasonings concerning matters of fact are based ultimately on the relation of cause and effect

                ·  I know my car is in the lot because I put it there, locked the door, etc.

·  I know the sun will rise tomorrow because I know the earth is caused to rotate by the sun-s gravitational pull

·  In know it will rain this weekend because of the clouds in the satellite picture, the wind conditions, and the relation between clouds and rain

·  All factual reasoning, says Hume, presupposes a causal connection between a present fact and what is inferred from it


      (present fact)                                                          (inferred event)

      Watch in sand                                                Someone has been here





             effect            presumed causal relation          cause


E.      Causality is Not Based on Reason

·  The next question that arises is, how is it that we have knowledge of cause and effect?

·  Does it come from the light of reason as Descartes contends? (PSR?)

·  Hume says that wherever it comes from, it does not come from reason

      ·  It is not known a priori

      ·  It does not derive from relations of ideas

·  Reflect on any causal inference, and you will see that the conclusion is never opposed to a contradiction

      ·  Denying conclusion doe not imply contradiction

      ·  Both conclusion and its denial are possible

·  Since all truths of reason have this trait, and causal inferences do not, they are not truths of reason


F.      Examples

1)      Smooth Pieces of Marble

·  No analysis of the pieces will reveal that they will stick together without appeal to a similar experience

·  Implies no contradiction to say that they will not stick together

2)      Gunpowder

·  No one could know gunpowder will ignite by just analyzing the substance without reference to analogous properties in other objects in experience

·  Implies no contradiction to suppose that it didn’t

3)      Loadstone

·  No argument a priori could be given to account for the attraction of magnets to


·  Implies no contradiction to suppose that it did not

4)      Billiard Balls

·  No analysis of a round ball will logically demonstrate that one ball will impart motion to another

·  No contradiction to suppose that the second ball will stand still, or move backwards, or both balls disappear


G.     The Need For Observation

·  The only way we know these facts about things is that we have observed similar events in the past

      ·  Take away past observations, and we are utterly befuddled as to what will happen

·  To prove this, Hume asks us to consider someone being presented with an entirely new object

·  He will not be bale to infer what effects it will produce, or what brought it about

            ·  If he can, it will be based on some analogous object he has had an experience of

·  Because all causal inference is based on observation, questions concerning ultimate causes cannot be answered, since we cannot observe them


H.     Math Will Not Help

·  Nor can a study of geometry or some mathematical science be able to discover ultimate causes (sorry Descartes)

      ·  Geometry only helps us to apply laws that we have learned through experience

·  We observe a regularity, represent it mathematically, then manipulate the symbols to compute future events

            ·  All this is done given the law from experience

      ·  Geometry does not discover, it manipulates what is given to it


Part II


A.     A Negative Answer

·  Hume has been led to questions about experience by inquiring into the nature of our knowledge of our world:


      Q:  What is the nature of reasoning concerning matters of fact?

      A:  It is founded on the relation of cause and effect


      Q:  What is the foundation of our conclusions concerning this relation?

      A:  Experience (not reason)


      Q:  What is the foundation of our conclusions concerning experience?

      A:  Not founded on reasoning or any process of the understanding

            ·  Neither demonstrated, nor intuited by the mind-s eye


B.     Admission of Ignorance

·  Hume pauses here and reminds us that we too often are unwilling to admit that we just do not know certain things

·  He is pointing his finger specifically at dogmatic philosophers who try to account for everything in their metaphysics

·  When we ask questions like, how doe we know causes, and what is the foundation of experience, we should realize our own limitations

      ·  Like Socrates, he admonishes us to be more modest in our pretensions

·  He reminds us that we cannot know nature-s darkest secrets, we can only infer patterns we see bits and pieces of the world through the senses

            ·  Why does bread nourish me?

            ·  Why is there centripetal force?

            ·  Why is motion communicated?


C.     The Nature of Causal Inference

·  Back to the question

·  In what way does experience support (provide evidence for) our causal claims?

·  The short answer is…

      ·  We appeal to post experience

·  Whenever we infer a cause given an effect, or vice versa, we implicitly or explicitly appeal to events that we have seen in the past

      ·  Egg break?

      ·  Window break?       It has in the past

      ·  Chalk fall?

·  This appeal to the past is legitimate if we make an assumption:

      ·  The future will resemble the past


D.     Inference and the Future

      ·  The question now is, how can we be sure that we can know this assumption?

            ·  How do we know that the future will resemble the past?

      ·  What kind of evidence could we provide for that claim?

·  How do we know that baseballs will break windows tomorrow as well as yesterday?

·  That eggs will continue to break?

·  That chalk will fall?

·  That the sun will rise tomorrow?

      ·  Does it come from reason - a priori - deduction?

            ·  No, no contradiction to suppose otherwise

            ·  The course of nature would change overnight

      ·  Does it come from experience?

            ·  Yes, but that does not help

                        ·  Basis of experience?  ŕ  future » past

                        ·  Future » past?  ŕ  experience

                        ·  This is a circular argument



E.      Hume-s Challenge

      ·  We still have no foundation for our knowledge of causal relations

      ·  We know it has something to do with experience

      ·  And that that has to do with the future resembling the past

      ·  We are left with skepticism

·  But that is not saying that there is no explanation; he s saying that he would like to know what it is!  How can we be absolutely certain of any causal inference?




not demonstrable                            from experience



                                                      future » past                 Circular



                              not demonstrable                from experience





A.     What is Evidence

·  One of the questions an epistemologist asks is, what is evidence?

      ·  More importantly, what counts as relevant and good evidence for a particular claim

      ·  If I claim that my car is in the lot, what evidence would I offer as support?

·  If I claim that a squared + be squared = c squared, what evidence would I offer as support?

      ·  Are these two pieces of evidence the same?  Of the same type?

            ·  Different kinds of claims require different kinds of evidence

            ·  That is, the way we justify a claim depends on the kind of claim it is


B.     Relations of Ideas and Matters of Fact

·  When Hume distinguished between relations of ideas and matters of fact, he had these sorts of questions in mind

·  When I make a claim about abstractions such as mathematical entities, I use evidence which derives from relationships between the concepts themselves

·  The relations between lines and angles of intersection provide the proof of the Pythogorean Theorum (logically indistinguishable ideas) 

·  On the other hand, when I make a claim about the physical world, my evidence must come from my contact with it - experience, observation

·  My contact with cars and their relations to places under certain conditions informs me that my car is (most likely) in the lot (logically distinct ideas)


C.     Causes and Evidence

·  Hume is concerned primarily with matters of fact, since they are directly relevant to our world

·  When we do not have direct evidence of the senses concerning a claim about the world, we must make an inference or inferences to justify the claim (provide evidence)

      ·  Reasoning always involves indirect evidence for an empiricist

·  Such reasoning, says Hume, is always based on a knowledge of cause and effect

      ·  Car is in the lot - How do you know? - you do not see it

      ·  I infer since I put it there and it is locked

·  So the evidence for claims about the world which cannot be based on direct observation must be causal - based on known relations between cause and effect


D.     How Do We Know Causes?

·  All knowledge of Matters of Fact are based either on direct observation (which Hume approves of) or on causal inference

·  The question is, if we are to trust our inferences, how do we know about cause and effect?

·  Can we infer it from reason a priori?

·  Can we know it from the analysis of concepts?

      ·  No, No, NO!!

      ·  No contradiction in presuming the opposite

      ·  No inference comes merely from analysis

      ·  No experience, no inference

·  It comes from experience - but what does that mean?