Test 1 Flashcards
What is critical thinking?
Thinking logically and rationally to arrive at the truth
Branches of Philosophy?
- Metaphysics
- Ethics
- Epistemology
- Logic
What is truth?
Aristotle: “Truth is saying of what is, that it is and of what is not, that it is not”
Simply = truth is where the statement made corresponds to reality
Subjectivism (relativism)
“X might be true for you, but X is certainly not true for me”
Relies on a singular objective reality accepted by arguing parties
Why think critically?
We want our beliefs to be true. Beliefs = map of reality, therefore want the map to be as accurate as possible.
Sources of Evidence?
- Sensory experience
- Memory
- Testimony
- Arguments
Arguments: Definition?
Sequence of statements (premise(s) and conclusion) intended to establish the truth of a statement
Include reasons and rely on our ability to evaluate evidence.
Difference between argument and rhetoric?
Difference b/w an argument and rhetoric is that a successful argument is an aid to understanding, while successful rhetoric produces a desired action
Principle of Charity
Should prefer good reconstructions of arguments, as the ultimate goal is truth.
Reconstructing an argument in a negative light to look like a bad argument is pointless, as you are no closer to knowing whether that claim is true or false.
Methods for evaluating evidence
- Look for a vested interest
- Watch for weasel words
- Question surveys
- Caution when dealing with surveys, percentages, numbers, anecdotes, stats and graphs
- Retain a sense of the big picture and proportion
Validity and valid arguments?
Validity = where premises guarantee the conclusion
Validity concerns the structure of the argument, not it’s truth or content (arguments can therefore be valid even if the premises are false)
Soundness
Sound arguments = valid arguments with premises that are actually true
All sound arguments are valid
Formal fallacies and types?
Fallacy = error in reasoning
Types:
Formal (form of argument)
Informal (content of argument)
Proper form:
P1: IF G, then F
P2: G
C: therefore F
G = anticedent, F = consequent (form of modus ponens)
Post Hoc Fallacy
“After this, therefore, because of this”
Claims that certain events are caused by previous events with no causal link
Informal Fallacy: Formalising argument?
Can’t identify an informal fallacy by formalising argument
Appeal to majority
Informal Fallacy
Uses popular belief in a proposition as evidence that the proposition is true
Ad Hominem
Informal Fallacy
Attacks trait of arguer to undermine arguers point
Appeal to Authority
Informal Fallacy
Uses claim of authority on facts that aren’t relevant to the argument to add credibility to the argument
False Dichotomy
Informal Fallacy
Presents a limited set of alternatives as though they are the only ones, when there may be others
Appeal to Ignorance
Informal Fallacy
Uses lack of evidence for the opposing proposition as evidence that the presented proposition is true
Straw Man
Informal Fallacy
Wilfully misrepresenting an opponent’s argument so that it looks superficially similar but it is much easier to attack
Slippery Slope
Informal Fallacy
Sliding from one likely event to progressively less and less likely events in ways that make the causal chain seem inevitable
Post Hoc (False Cause)
Informal Fallacy
Claiming that certain events are caused by previous events with no causal link
Genetic Fallacy
Informal Fallacy
Assessing the truth of a claim by it’s origin or history, as opposed to its relevance or truth
Argument from tradition
Informal Fallacy
Uses the fact that people have historically believed a proposition to be true as evidence that the proposition is true
Begging the Question
Informal Fallacy
Uses a proposition as a premise in an argument that is intended to defend that very proposition
Hasty Generalisation
Informal Fallacy
Draws a general conclusion on the basis of insufficient evidence
Equivocation
Informal Fallacy
Uses same word or phrase to mean two different things
Red Herring
Informal Fallacy
Changing what is at issue in an argument in a way that distracts from the issue at hand
Loaded Question
Informal Fallacy
Asks a question informed by implicit assumptions
For example, the question “have you stopped mistreating your pet?” is a loaded question, because it presupposes that you have been mistreating your pet.
Composition/Division
Informal Fallacy
Uses the attributes of individuals as evidence that the group has that attribute, or vice versa
e.g individual —-> group (composition)
group ——> individual (division)
Moving the goalposts
Informal Fallacy
Set a standard for evidence and once this is met, raising the standard again or setting it somewhere else
No true Scotsman (Appeal to Purity)
Informal Fallacy
Setting up criteria for a certain argument, then redefining the criteria to exclude certain outcomes.
Heuristics
Cognitive shortcuts that enable us to make judgements quickly and reliably
Bias: Confirmation
Look for/notice/ remember evidence that confirms our opinions/favoured theories whilst ignoring disconfirming evidence
Bias: Negativity
Attend more closely to & place more weight on negative evidence than positive evidence
Bias: Availability
Tend to give a lot of weight to evidence that is easily accessible
Bias: Anchoring Effect
Let initial value affect our appreciation of subsequent values
Bias: Neglecting Probability
Tend to be unmoved by information about the probability of various kinds of events. Therefore, tend to overestimate risks of relatively safe activities while underestimating dangerous ones
Bias: Observational selection
Begin to notice something that we didnt notice before and assume that the frequency must have increased
Bias: Status Quo
Fearful of change even when the status quo isnt great
Bias: Bandwagon Effect
Tend to follow the crowd, sometimes unconsciously, often even when the crowd is wrong
Bias: False Consensus
Overestimate how typical or normal we are and assume that most others think like us and believe what we believe
Bias: Current Moment
Not good at imagining ourselves in the future therefore discount future costs and benefits
Bias: Hindsight
Tend, after something has happened, to believe that we knew that it would happen all along
Bias: Actor-observer
When asked to identify causes of an event, we either:
(a) If involved (as actors), tend to identify the situation as the cause
(b) Merely saw it (as observers), tend to identify the actor as the cause
Bias: Ingroup
Place more weight on the views of members of our immediate group than those of outsiders
Two types of AI?
Weak: AI focussed on a narrow task such as winning a game of chess
Strong: AI where the machines intellectual capability is equal to a humans
Biases in AI are caused by?
(a) Bad Data
(b) Biases transferred from humans to AI
Common tech behind AI include
- Regression based systems
- Deep learning/artificial neural network
- Expert Systems
Automation Bias
Tendency to disregard or not search for contradictory information in light of a computer-generated solution that is accepted as correct
AI Bias; Historical & Examples
Often build and train algorithms that are used to make predictions with historical data therefore inherent past bias
Examples:
Predpol (Feedback Loop Issue)
- kept sending police to ‘recorded crime’ areas therefore always kept finding crime there = feedback loop
COMPAS
- assesses risk of offender re-offending. However, used historical data which was skewed unfairly against black individuals
Bias: Interaction and Examples
Chatbots = learn how to answer questions based on interactions they have with public
Microsoft Tay; input data taken from twitter conversations which was mainly racist, misogynistic individuals
AI Bias: Sample
Data used to build AI is not representative of space within which the model will operate
E.g. Google Photo Disaster
AI Bias: Input Selection
Choose factors as inputs which are biased. In case of risk prediction algo’s, often lead to minority populations being disadvantaged.
E.g. Allegheny County Algorithm
AI; Options to solve issues
- Human in the loop
- Right to an explanation
- Choice to not be subject to an algo
Independent Reasons
Reasons that provide evidence on their own`
Dependent reasons
Reasons that only provide evidence when taken together
Enthymemes
Arguments with unstated premises
Rhetorical Devices
Sometimes premises or conclusions are presented as questions or commands
