The Missing Link in Hume's Account of Causal Reasoning and Causal Belief


Book 1, Part 3:


Section 14.  The Idea of Necessary Connection


Para. 1:  Custom determines the mind to make a transition from a cause to an effect (or vice versa).  The necessity is an impression of reflection.


Para. 4:  The circle of equivalents:  cause, efficacy, agency, power, force, energy, necessity, connexion, and productive quality.


Para. 5:  What is the source of the idea of necessity?  Not reason. 


Paras. 6-11:  The many futile attempts by other philosophers to solve the puzzle. 


Here Hume especially criticizes the occasionalists (represented by Malebranche), as explained by Bell in his article in the Cambridge Companion.








The Paradox of Non-Existent Ideas

(Recall Hume’s discussion of substance)


        How can Hume claim that we do not have the idea of X without using that very idea?


Let X = power


        “If we have an idea of power, we may attribute power to an unknown quality.  But as ‘tis impossible, that that idea can be deriv’d from such a quality, and as there is nothing in known qualities, which can produce it; it follows that we deceive ourselves, when we imagine we are possest of any idea of this kind, after the manner we commonly understand it.”[T1.3.14.11]


What does “after the manner we commonly understand it” mean?  If we can’t have the idea of power as a hidden quality of objects, how could we commonly understand it that way?


        “We never have any impression, that contains any power or efficacy.  We never therefore have any idea of power.” [T1.3.14.11]


        How can he tell us what idea we don’t have?






Para. 12:  Can we get the idea of cause from the example of our will?


        Hume says that we do not perceive the causal power of the will.  He just asserts that we come to believe in the will’s causal powers by the same process that we acquire any beliefs about causes. 


Could this be correct?  Did you come to believe in the causal powers of your will by noticing  a constant conjunction between decisions to do X and the corresponding actions?  Could you have?


















Para. 13:  Here Hume makes an assumption that is not warranted.  He assumes that the only alternative to his account of necessity is the necessity of demonstration.


Paras. 15-21:  A recapitulation of the account of causal inference.  Here Hume’s goal is to convince us that the only idea of cause is the one that he is able to explain.


Para. 22:  The source of the idea of necessary connection in a determination of the mind. 


Para. 23:  The analogy to the necessity of reason.


Para. 26:  Objections.


Para. 27:  What does Hume mean that he is prepared to allow qualities in material and immaterial objects with which we are utterly unacquainted?  How could we have an idea of any such properties?


Para.  29:  Hume applies his theory of cause to his causal theory of the mind.  Hume’s own theory of the relation between impressions of sense and ideas is the result of constant conjunctions in the past.


Para. 31. 


Hume's position restated.





The Two Definitions of “Cause”



Para. 31:  The philosophical relation:  Cause = “An object precedent and contiguous to another, and where all the objects resembling the former are plac’d in like relations of precedency and contiguity to those objects, that resemble the latter.”[T]


        The natural relation:  “A Cause is an object precedent and contiguous to another, and so united with it, that the idea of the one determines the mind to form the idea of the other, and the impression of the one to form a more lively idea of the other.”[T1.3.14.31]


















Para. 32:  One kind of cause:  All causes are efficient causes.


Para. 33:  One kind of necessity:  Moral and physical necessity are the same.  (What about logical necessity?)


Para. 36:  Hume’s parting maxim: “We can never have any reason to believe that any object exists, of which we cannot form an idea.”[T1.3.14.36]























Section 15.  Normative Hume's Rules to Judge of Causes and Effects [Rules for good causal reasoning, with a small ‘r’]


1. Contiguity in space and time


2.  Temporal priority.


3.  Constant conjunction.


4.  Same cause, same effect.


5.  Like effects imply(?) like causes.

Is this true?


6.  Different effects must be due to differences in causes.


7.  In some cases, but not all, increases and diminutions in causes produce corresponding increases and diminutions in effects.


8.  A time separation between a purported cause and an effect shows that the first is not the sole cause of the effect.  [This is a corollary of #1.]








Question:  If the mind works by causal laws, how can there be any norms for good causal reasoning?


Hume’s answer:  The mind’s reflection on itself can change the force and vivacity of its ideas.  If we notice ideas that violate the rules, this very noticing will diminish their force and vivacity.

























Sect. 16:  On the reason of animals


What kind of reasoning do beasts exhibit?  Do they need the faculty of Reason to do it?


Do beasts have knowledge?  Do they have probable belief?


What does Hume mean by "instinct of animals"?  How does animal instinct confirm Hume's view that belief in cause and effects does not involve Reason?


What does Hume mean when he says “reason is nothing but a wonderful and unintelligible instinct in our souls”[T1.3.16.9]?  Does he mean Reason with a capital “R”?
















Puzzles for Hume’s Account of Causal Inference



(1) Pavolovian conditioning and common causes.

Why don’t we infer that the sound of a gunshot causes death?


(2) Causes and effects experienced simultaneously (e.g. a pin prick).


(3) Countervailing causes:  two balanced teams in a tug of war.

On Hume’s account, doesn’t this possibility show that nothing is truly a cause.  (See T


(4) Hidden causes.  

What about things that can only be known by their effects  (e.g. earthquakes)?  [Think of Hume’s account of substances.]

What about hidden mental causes (e.g., memory that is not conscious or the rule that accompanies an abstract object idea or habit).  Have you ever experienced an unconscious memory or an abstract idea rule or a habit?









Hume's Skepticism About Bodies (Physical Objects)



A.  The Ideas of Existence and External Existence


Book 1, Part 2, Section 6:


There is no idea of existence separate from our ideas of things (as existing). 


Question:  How can we have the idea of something (e.g., a unicorn) that does not exist?


Key:  Can we have the idea of external existence (of something existing outside the mind)?


Another example of the Paradox of Non-Existent Ideas.















Book 1, Part 4:


Section 1:  Skepticism with Regard to Reason


Reason is a kind of cause.  What is the natural effect?


Hume argues that we must adjust our confidence in the results of our reason based on our past experience of error.  For example, in doing a complex math problem, we must adjust our confidence in the conclusion based on the frequency of mistakes we have made in the past on similar problems.  How can we estimate the frequency of mistakes in the past?


This argument is a landmark in epistemology.  It marks the beginning of the end of pure a priori justification.


Hume then gives a fallacious argument for why all probability (belief) must eventually degenerate into improbability (unbelief), and thus why Reason cannot avoid skepticism.  We will set that argument aside.


Having disposed of Reason, Hume is ready to turn his attention to what most interests him, probable belief based on perception, memory, and the imagination.







Section 2:  Skepticism with Regard to the Senses


In this long section, Hume is going to follow out the implications of his causal theory of the mind. 


para. 1:  “We may well ask, What causes induce us to believe in the existence of body? But ‘tis in vain to ask, Whether there be body or not?  That is a point, which we must take for granted in all our reasonings.”[T1.4.2.1]

[See, also, para. 38]


This entire section is not to convince us to stop believing in bodies, which would be vain, but to clarify why it is that we believe in bodies.



para. 2:  What does Hume mean by "body"?

the concept of body can be divided in two parts:  (a) continued existence and (b) distinct existence.  Distinct existence itself has two parts:  (i) external and (b) independent.


The puzzle:  Earlier, Hume had argued that we cannot have an idea of external existence.  How can we have an idea of continued existence or distinct existence?



The three possible causes of our belief in body: 


paras. 4-13:  It's not the senses. 


(1) We could not get the idea of continued existence from the senses. 


(2) What about external existence?


Hume’s example of the paper beyond his hand, with the fields beyond (para. 9).


Note that here he appeals to other philosophers for authority, because common sense would say that we do perceive objects as external to us.


What about independence?


In paras. 12-13, Hume uses the example of secondary qualities (e.g., color) as a model for all sensory properties.  He takes it for granted that physics has shown that objects are not colored, but that the colors are in our minds.  He generalizes that argument to all sensory properties.


para. 14:  It's not Reason.  Why not?

Even children and peasants believe in bodies.   Their beliefs could not be based on Reason.

The main argument is that beliefs about bodies are probable, so they could not be the result of Reason.


paras. 15 ff.  It is the imagination. 


How does the imagination give us the idea of continued and distinct existence?  [This is a deep question.]


It cannot be explained by force and vivacity.  Why not?


Paras. 18 and 19:  The answer: constancy and coherency.

What is constancy?

What is coherence?


In his explanation, Hume reverses the order: 


paras. 20-22:  What is responsible for our belief in distinct existence?  Coherence.

The example of Hume and the porter.

The example of the letter.


Note that Hume here seems to be introducing a new kind of causal inference, not based on constant conjunction (para. 21).  See also paras. 42-44 and 47.


Hume acknowledges that it is a new kind of inference.  Why can’t it be subsumed by his account of causal inference?


The tension is becoming intense:  How could the imagination infer an idea (body) that is not derived from an impression or made up of ideas that are derived from impressions?


paras. 23-24:  What is responsible for our belief in continued existence?  Constancy.


On Hume’s account, we produce the illusion of an idea that continues when it is unperceived.  How could we ever form that idea?


Pay attention to what Hume calls “illusion” or “fiction”.

In para. 29, time is called a “fiction of the imagination”.


Hume’s experiment of shutting his eyes (para. 35).


Para. 39:  Hume’s theory of the mind as a “heap or collection of different perceptions, united together by certain relations”.


Para. 41:  What is the source of the belief in continued and distinct existence?


Paras. 42-44:  Propensity to feign continued existence bestows vivacity on the “fictions”.


How do we know it is a fiction?


Para. 45:  Hume’s famous experiment of double perception.

This experiment refutes the vulgar view of bodies.  (What is the vulgar view?)


Para. 46:  What is the philosophical view?  The opinion of double existence.





Paras. 47-48:  The philosophical view could not be endorsed by Reason or the imagination.

Why not the imagination?  There is no causal inference from perceptions to bodies.  [In para. 51 Hume attributes the belief to the imagination.  Also, recall Hume’s discussion of coherence in paras. 20-22.]


What is Hume's conclusion?

Why can no one believe it?


But if we could never have an idea of "body", how could we believe in bodies?


Para. 51:  the difference between reflection and instinct.


Does he think that his philosophy will stop us from believing in bodies?


How could we ever start believing in something of which we can have no idea?


Paras. 56-57:  Hume began the Treatise as a vindication of common sense.  The final tension is that now he finds that on reflection, he cannot endorse common sense.








Hume refers to constancy and coherence as “trivial qualities of the fancy” (para. 56).  But it is more plausible to think that what he has really shown is that they are a powerful kind of reason for believing in bodies, thought of as “constructs” rather than “fictions”.


Digression on how Quine repeated Hume’s mistake.  Both recognized the importance of coherence relations in reasoning, but did not have any way of explaining how such reasoning could be possible, because of their simplistic causal theories (Hume’s reduction of thought to customary connections between perception makes reasoning a kind of conditioning process.  Quine adopted a more extreme conditioning theory, behaviorism, which replaced perceptions with patterns of stimulation of our sensory receptors.


Key claim:  “’tis impossible for us distinctly to conceive, objects in their nature any thing but exactly the same with perceptions.”[T]


Why is it impossible?


What is the remedy to these skeptical doubts?  “Carelessness and inattention” (para. 57).  [Instinct? What is that?  It is the laws of the mind, especially the laws governing the imagination.]