Rationalism v Empiricism From where does knowledge come

Rationalism –v- Empiricism From where does knowledge come? MRes Philosophy of Knowledge (slides available on Moodle and at http: //cfpm. org/doctrain)

Some Key Foundational Ideas Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-2

(Potted) History of the “Western Liberal” Tradition of Thought • Start usually attributed to culture of Ancient Greeks from around 600 BCE • Taken up by Romans (some aspects) • After Roman empire collapsed, was maintained/developed in the Islamic World • Later re-imported to Western Europe • At different times nutured in different European Countries • Now in many countries across the world Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-3

The Original Greek Context • Small, independent but affluent “city states” • Where the citizens discussed court cases, and some decisions collectively • (the “citizens” did not include women, slaves, outsiders or children) • Thus rhetoric and argument were important • This was a social process • The outcomes of these discussions were important – they had real consequences Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-4

A picture of reasoning we get from Ancient Greece • Reasoning is (largely) a public activity involving rhetoric and debate • An effort to distinguish (and exclude) bad arguments (e. g. identifying “fallacies”) • Good argument results in good decisions • Argument seen to lead from known (i. e. agreed) truths to conclusions… • …which may be new or used to show other beliefs are wrong (or inconsistent) Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-5

This activity is abstracted to “Reason” and “Rationality” • These are ideals – Not necessarily what people actually do – They acquire a normative flavour • Major questions are then: – What is the role of Reason (as opposed to perception, action etc. ) in particular as to its relation to knowledge? – What is Rational (in contrast to irrational)? • Can be seen as a search to escape the contingencies and particularities of the observed world… • …to more general/fundamental/etc. truth Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-6

Deduction • An activity – something one does • Generally what happens in an argument • Can be statistical, logical, mathematical, computational, linguistic • Finding the necessary consequences of other propositions • An analytic “unfolding” or “working out” of what is already known • Produces a new form of old knowledge • If we know A and A B we can deduce B • Can be pretty watertight – true by its nature Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-7

Some examples of Deduction • By counting votes in a room, and the number there concluding that some abstained • Calculating the average of a set of numbers • Working out the answer to a Sudoku puzzle • Finding, by elimination, where a fault in some wiring lies • By looking at a timetable, determining the possible busses that one could take • By looking at the rules, determining how much extra income tax one will owe Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-8

Induction • Another activity – something one does • Usually spotting patterns or facts from observations or data • In other words something is learned • Finding generalisations from evidence • Results in new knowledge • S 1 was W, S 2 was W, S 3 was W, …. All Si are W • Obviously fallible, i. e. it could be wrong Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-9

Some examples of Induction • Having observed many swans and that they are all white, that all swans are white • Since, in the past, every time one has gone to sleep one later wakes up that this will also happen the next time one sleeps • From observation that people get more wrinkles as they get older • Daffodils flower later in the year than snowdrops • There is more traffic on the roads between 4 pm and 6 pm than between 3 am and 4 am Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-10

Exercise on Induction and Deduction • In small groups, choose an example from the sheet • Try and work out – what mix of deduction and induction is being used – how and where each of deduction and induction are applied – whether the conclusions are more deductive or inductive • If you have time pick another one etc. Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-11

The necessary – contingent distinction • Necessary truths – Statements that have to be true – For some reason it is not possible that they could be false – e. g. Sentient beings exist • Contingent truths – Statements that just happen to be true – If things were different they might not have been true – E. g. The UK has a Reigning monarch Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-12

Exercise on the necessary – contingent distinction Which of the following are necessary and which contingent truths? • Cows have four legs • 1+1=2 • We exist • If Aristotle is a man, and all men are mortal, then Aristotle is mortal • The world is composed of objects • Meaning is not from the world, but something we create and project onto the world • Either Trump or Clinton will win the 2016 Presidential Election Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-13

The analytic – synthetic distinction • Analytic truths – True by definition or deduction – Are necessary – Abound in mathematics or logic – e. g. All bachelors are unmarried • Synthetic truths – True of the world – Are usually contingent – Abound in natural sciences – e. g. The Earth orbits the Sun Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-14

Exercise on the analytic – synthetic distinction Which of the following are analytic and which are synthetic truths? • All cats are mammals • All mammals feed their young with milk • No mammals lay eggs • All swans are white • The average of a set of numbers is not greater than their maximum • All numbers can be expressed as a fraction • All humans have rights Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-15

The a priori – a posterior distinction • a priori knowledge – What one knows before taking into account observations or evidence – May include necessary/analytic truths, assumptions, given facts, etc. • a posterior knowledge – What one knows after taking into account observations and evidence – May include laws and explanations of natural or social phenomena Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-16

Exercise on the a priori – a posterior distinction Which of the following are a priori and which a posterior truths? • What words mean • That I will be able to recognise your emotions from your facial expressions • That an experiment worked • That experiments should be designed to reduce extraneous factors • That arguments have conclusions and premises • 1+1=2 Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-17

The context of discovery – context of justification distinction • The context of discovery: – The situation/context where an item of knowledge is discovered or hypothesised – When and how knowledge is learnt • The context of justification: – The situation/context where the knowledge is justified, established or verified – When and how knowledge is established as reliable e. g. Fleming discovered penicillin when he accidentally let a culture be contaminated by mould but … it was justified as knowledge as the result of further experiment and observation by many Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-18

The Idea of Causation Event A Event B One event (A) causes another (B) if • B always follows A • A is necessary to B occurring • And (generally) there is some mechanism connecting A to B E. g Does smoking cause Cancer Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-19

Exercise on the Idea of Causation Which of the following could be examples of causation? • Smoking causes cancer • Being black makes me proud • Adding 1 to 1 makes 2 • Taller people tend to be paid more • Person A will die because they are seriousl ill • Persona A will die because they are doomed • Living in a poor area of Manchester makes you less likely to go to university Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-20

The Rationalist/Empiricist Debate Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-21

Francis Bacon’s Two Ways to Truth: from the general to the particular “There and can be only two ways of searching into and discovering truth. The one flies from the senses and particulars to the most general axioms, and from these principles, the truth of which it takes for settled and immovable, proceeds to judgement and the discovery of middle axioms. And this way is now in fashion. … Reason General Principles Contingent facts and observations Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-22 Specific Truths

Francis Bacon’s Two Ways to Truth: from the particular to the general …The other derives axioms from the senses and particulars, rising by a gradual and unbroken ascent, so that it arrives at the most general axioms last of all. This is the true way, but as yet untried. " Francis Bacon (1620), First Book of Aphorisms Reason Contingent facts and observations Specific Truths etc… Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-23 General Principles

Rationalism • A Philosophical position – something one believes – about where knowledge comes from • Knowledge arises from reasoning • Is the position that the way to knowledge is from the general to the particular • Requires some general a priori truths which it views as necessary (usually) • Characterised by deduction • The general principles gives meaning to the observations by relating them • E. g. Descartes’ “I think therefore I am” Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-24

Why be a Rationalist? • Data alone does not tell you anything • Knowledge and reason seem intimately connected • To claim knowledge you need to justify its truth and this requires reason • All knowledge requires a host of other assumptions upon which it rests, the relationship between these requires reason • Without reason there is nothing but ‘brute facts’ and undistinguished perceptions • To have meaning requires some structure Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-25

Empiricism • A Philosophical position – something one believes – about where knowledge comes from • Knowledge arises from observation • Is the position that the way to knowledge is from the particular to the general • Requires some particular a posterior truths (perceptions) which are contingent • Characterised by induction • The general principles arise from the process of relating observations • E. g. Galileo’s “All truths are easy to understand once they are discovered; the point is to discover them” Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-26

Why be an Empiricist? • Where else could knowledge come from? • If no evidence was involved (even way down the line) is not all reason circular? • Our assumptions and conceptual frameworks are frequently simplistic • We are very good at deceiving ourselves with our own clever thought! • The world is far too complex for us to conceptualise – data presents this to us Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-27

Example: please assess • • What evidence is presented? What kinds of evidence does he use? What assumptions does he rely upon? What is the nature of the rationality that the author presents? • What arguments does he present? • What does his argument seek to show? • Do you find it convincing? Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-28

The Example: more issues • Is the author a Rationalist or an Empiricist? • Does the author rely on a priori truths or a postiori evidence? • Is the author claiming his arguments are necessarily true or only contingently true? • Has the author go to his conclusions using deduction or induction? Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-29

An Example of a Rationalist Theory – Rational Choice Theory Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-30

Basic Idea of Rational Choice Theory (RTC) • Humans are (imperfectly) rational • That is, their behaviour can be explained (modelled) by assuming they are basically rational at least as a starting point • The conception of rationality here is a generalisation of economic rationality (constrained optimisation) • E. g. Smith’s “invisible hand” balancing supply and demand via price changes Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-31

Principles of RCT • Resourceful: man can search for and find possibilities, he can learn and be inventive; • Restricted: man is confronted with scarcity and must substitute (choose); • Expecting: man attaches subjective probabilities to (future) events; • Evaluating: man has ordered preferences and evaluates (future) events; • Maximizing: man maximizes (expected) utility when choosing a course of action Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-32

About RCT • The idea is that if you accept those principles, this leads to a picture of enumerating possibilities and choosing the best of these • RCT is a rationalist theory (as well as being about rationality) because… – It does not really have an evidential basis – But rather is defended on the basis of what can be explained on “first principles” • Despite its lack of empirical grounding it has been very influential, since it presents a picture of what rationality “should” be Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-33

An Example of an Empiricist Theory – Administrative Behaviour Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-34

Its Origin • Herbert Simon attacked RCT (and similar) as being “armchair economics” since it could have been invented sitting in an armchair rather than observing anything • Instead he observed people within offices to see how they made decisions and came to a different hypothesis as a result • He called this “procedural rationality” to contract it with “ideal” (what he called “substantive”) conceptions of rationality (such as RCT) Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-35

Procedural Rationality (PR) That (within constrained and familiar environments like offices): • People tended to have developed sequences of actions, procedures, that they followed if all was normal • They did this regardless of the existence of “better” alternatives… • …as long as it was “good enough” • That is achieved a minimum threshold in terms of quality (it was “satisfactory”) Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-36

About PR • It was born from an empiricist approach, coming from observational evidence rather than reasoning from principles • Oddly had far more influence within AI than in the social sciences • Has later been (incorrectly) summarised as “satisficing” – which was the same as RCT but where a search through alternatives was terminated when utility met a given threshold Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-37

A brief account of some of the arguments in this area Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-38

Recap of Induction Supposed structure: – The 1 st swan is white – The 2 nd swan is white – … etc. – Therefore all swans are white • • Relies on there being observable patterns Produces (fallible) generalisations A source for hypotheses and theories A natural thing to do Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-39

The Problem of Induction • Hume A Treatise Concerning Human Understanding • Although one does repeatedly observe a particular conjunction (or sequence) of events… • …this never guarantees that this will always be the case. • Thus there are no causal laws Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-40

The Inductive Justification of Induction • The argument – Induction worked in case 1 – Induction worked in case 2 – … etc. … – Therefore induction works in all such cases • • A self-referring and self-justifying argument But if it is false it does not justify itself Is it supported by the evidence? What is the scope of the cases? Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-41

Some difficult questions about induction • What are these facts? – Are they states of the world? – Are they statements in language? – Are they something else (propositions)? • How do we select these facts? • Why did we look for these facts? • What conceptual framework did we use to construct our generalisations about them? • What background assumptions are there common to all these facts? Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-42

Popper’s Falsificationism • Induction never proves anything • Hypotheses can only be disproved by observing a counter-example (a black swan) • We rely on hypotheses more as they survive attempts to disprove them • If there is constant innovation of hypotheses and attempts to disprove them then knowledge will progress • Hypotheses that are not amenable to being falsified (unfalsifiable hypotheses) are dubious Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-43

Comments on Popper’s Falsificationism • History of science does not fully support it (e. g. Michelson-Morley experiment) • How does one know whether the counterexample shows the main hypothesis is wrong or merely an auxiliary assumption? • Marks a switch from the context of discovery to the context of justification • Results in an evolutionary picture of the development of knowledge (evolutionary epistemology) Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-44

Lakatos’ Core and Protective Belt • Research programs as key entities – These have a core of fundamental frameworks, methods and assumptions that characterises them – And a belt of less fundamental hypotheses, observations, techniques • In the face of counter-examples research programs change things in the belt and preserve the core • Some programs are more successful than others (the “degenerate programs”) Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-45

Exercise: considering programmes For your own field/area of work decide: • What key deductive techniques does it use/rely upon (are there any not used)? • What key inductive techniques does it use/rely upon (are there any not used)? • Do these tend to be used in any particular order, if so what order? • What are the core assumptions/tenants of your field (i. e. if you don’t hold to them you do not really belong in this field)? • Are claims effectively falsifiable, if so how? Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-46

In the Social Sciences • One is usually dealing with meaningful behaviour • Meaning is (almost always) an a priori given but not always agreed upon • The context (or scope) of the induction or falsification is of great importance • Falsification is difficult since it is easy to adjust the belt to protect any hypotheses • Coherency with other thought often as important (to academics) as evidence Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-47

Summary of Session • Both reason and evidence are needed for good generalisations about the world • But how these are combined is important • When and how to use/mix induction and deduction is still a big issue in all fields • Care and awareness are needed with any a priori assumptions and frameworks… • …although ultimately these are unavoidable • Key decisions are what to do if some evidence seems to conflict with a hypothesis – beware! Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-48

The Reading List • Please read “What is thing called Science? ” by Chalmers to be introduced to some of the key ideas • Pick some of the other things from the reading list as seems appropriate • Read anything in your field that touches on the philosophy involved • Come and ask me for suggestions if you are stuck • Keep reading – even if difficult to start with Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-49

The End Experiment escorts us last His pungent company Will not allow an Axiom An Opportunity Emily Dickinson (1830 -1886) (as usual slides etc. at: http: //cfpm. org/mres) Rationalism –v- Empiricism. Philosophy of Knowledge, Bruce Edmonds http: //cfpm. org/doctrain slide-50