Wednesday 12/7: Last discussion of the semester! Analogies, plus odds & ends

Announcements:

• Quiz #6 posted this afternoon (Weds. 12/7), due by midnight!
• Final project due by midnight Sunday, 12/11
• Group component
• Names!
• Individual component
• Peer evaluation component - for those working in groups
• Email me or come see me with questions
• Final exam (replace your lowest quiz grade) Friday, 12/16

Plan for today

• Focus on arguments from analogy
• Basic structure
• Evaluating premises and argument strength
• Work on examples in groups
• End of class: Course evaluation for recitation component of class
• Volunteer to return completed evaluations and pencils to Shatzel Hall 305?
• If the office is closed for lunch, slide the envelope under the door and leave pencils outside 

Takeaway points for today:

• Familiarize yourself with the structure of arguments from analogy and be able to a) critically evaluate the premises and b) assess the arguments' overall strength
• For the quiz tonight, be prepared to answer questions on arguments from analogy, and:
• Various types of scientific trials
• The "drug development pipeline," esp. the relationship between phase of testing and strength of evidence, generalizability, and so on
• The importance of blinding (blind/double-blind studies)
• The difference between subjective and objective measures

Arguments from Analogy

Argument from analogy = def. A type of inductive argument in which one infers on the basis of known/observed similarities between two things (or classes of things) that those two things (or classes of things) share another, unknown similarity. 



Basic structure

P1) A and B are alike with respect to w (and x and y...)*
P2) A is (or has) trait z
P3) Being (or having) trait(s) w (and x and y...) is relevant to being or having trait z
C) B also is (or has) trait z

* The person making the argument from analogy won't always make this list of similarities explicit. So, as discussed in lecture, you may have to infer the contents of P1 yourself.
 
Example:

Kevin is a philosopher and he watches Black Mirror. It follows that Liz probably also watches Black Mirror because she's a philosopher too.  

Formalized: 

P1) Kevin and Liz are alike insofar as they are both philosophers
P2) Kevin watches Black Mirror
P3) Being a philosopher is relevant to (aka, predictive of, associated with, etc.) watching Black Mirror
C) Liz probably watches Black Mirror, too 



Critically evaluating arguments from analogy

• Like other inductive arguments, arguments from analogy can be weak, medium, or strong (logical force). 
• To determine the strength of an argument from analogy, we evaluate each premise in turn for acceptability. 
• In addition, to evaluating the premises we must consider the extent and relevance of the differences between the two things being compared. 
• You can formulate counter-arguments to and reductios of arguments from analogy. We'll discuss these concepts in class on Friday. 

Evaluating the premises of an argument from analogy

P1

1. Number of instances compared in P1
In general, more instances compared = stronger analogy

2. Variety of things compared in P1
In general, more variety = stronger analogy

3. Number of shared characteristics in P1
In general, more shared attributes = stronger analogy


P2

Not much to do here besides assess the statement for truth or falsity.


P3  

4. Relevance of shared characteristics in predicting the z property.
In general, more relevance = a stronger analogy. 



5. Revisit P1 and P3: Consider the number of dissimilar characteristics (vs. number of similar, P1) and the relevance of those different characteristics (vs. relevance of similar, P3) in predicting the z property.
In general, fewer relevant differences between A and B = a stronger analogy.


Step 5 is especially important. Your brain latches on to similarities between things (partly due to confirmation bias). But in order to critically evaluate arguments from analogy, you must also consider the relevant differences between the things being compared. So, force yourself to list the differences! Then, consider whether those differences are relevant to predicting whether something has property z.

Applying the tests to arguments: a systematic method

• Formalize the argument from analogy 
• Remember, the contents of P1 are usually not explicit
• Evaluate P1 and P3 with respect to #1-4
• #5: Make a list of noteworthy differences between the two things being compared
• Compare the list of similarities in P1 to the list of differences you've created with respect to two criteria:
• Number of noteworthy similarities vs. number of differences <-- Less important
• Relevance of those similar vs. different traits to predicting property z (P3 vs. your list of differences) <-- More important
• Finally, based on your assessment of the premises, determine whether the overall argument is weak, medium, or strong

Practice (groups)

a) Put the example in standard form for an argument from analogy
b) After formalizing the argument, make a list of differences (non-shared traits) between the two things being compared 
c) Evaluate the premises of the argument using the method described above
d) Determine whether the argument from analogy (overall) is weak, medium, or strong w/r.t. logical force 

  
Do questions from the "More Practice" section of Monday's lecture 

 

Course evaluations:

• Last 10-15 min. of class
• Recitation component of the course only (you will evaluate the lecture on Friday)

Thanks for all your hard work this semester, and good luck with finals!

  

 

Wednesday 11/30 Discussion: Pseudoscience, Scientific Reasoning, and Critical Thinking

Announcements:

-Two weeks of class left! 

• Like this class? Recommend it to a friend for Spring!

-Final project due Sunday 12/11 (early submission okay/encouraged)

• Posted on Canvas
• Groups of 3-5, or work individually*

* If you decide to work on the project individually you must notify me in advance. 

• Start of class: find groups for people who don't have a group (esp. those absent on Monday) and want one.
• Reminders:
• Peer assessment figures in grade (groups)
• There are group & individual components to the project
• The individual component must be done and submitted individually even if you're working in a group on the rest of the project 

• START NOW!!!1!
• Come see me with questions 

-Final exam Friday 12/16 (exam schedule online)

• Mandatory attendance
• Replace your lowest quiz grade

  

Plan for today: 

-Return Exam #2 

• Remarks 
• Review?
• Come see me with specific questions/concerns

-Discussion 

• Class findings:
• Pseudoscience & scientific reasoning homework
• Comparative evaluation of evidence for efficacy of supplements/herbal remedies/alternative medicines
• Create your own fad diet
• *The fake news epidemic*
• Critically evaluating scientific findings
• Elements of a good clinical trial
• Things to look for when evaluating a scientific study
• From Monday (11/28): Phases of drug testing and strength of evidence
• Critically evaluating claims about scientific findings
• See criteria above, and consider:
• Second-hand reporting
• Misinterpreted results
• Sensationalized headlines 

-Takeaway points:

• How to evaluate scientific studies and findings (features that make them more vs. less credible)
• Why we should be careful of second-hand reports of scientific findings, especially from non-experts/the media
• How the Interwebs facilitates the spread of pseudoscience (and false information, in general)
• Be a critical and skeptical reader!   

Pseudoscience and scientific reasoning


FOR REVIEW. Not the focus of discussion. However, you need to be familiar with these concepts in order to meaningfully participate in discussion.


Elements of a Good Clinical Trial

1) Control Group
2) Double blinding
3) A third treatment group (this concept may not be familiar)
With some interventions we know that just about any intervention will be better than no intervention - and some intervention may be ethically obligatory (e.g., in cases of suicidal behavior). 
So, rather that having only no treatment vs. treatment, you need to measure new treatments vs. the current standard of care/drug. 
4) Objective outcome measures
5) Random sample (e.g., avoid self-selection).
6) Placebo control

Terminology
 
1) Ideomotor effect
2) Post-hoc (after-the-fact) rationalization
3) Placebo
4) Nocebo

Things to Look for When Evaluating a Study

1) Effect size
2) Duration of effect
3) Type of study 
4) Funding
5) Reporting
6) Context in literature
7) Meta-analyses (this concept may not be familiar) 
Meta analysis = def. A study that combines similar studies on a topic to evaluate the overall trend. 
The rule of thumb for evaluating meta-analysis is "garbage in, garbage out." 
In other words, if most of the studies included in the meta-analysis are of poor quality then this will be reflected in the conclusion of the meta-analysis. When evaluating meta-analysis always read the section on inclusion criteria.
8) Replication
9) Impact number of journal (this concept may not be familiar)
Impact number = def. A journal's credibility rating
-Google it!
-Credible academic journals are refereed, peer-reviewed, cited by experts, etc...
-High impact #
versus
-"Academic journals" that are really "pay to publish" journals
-Low (or nonexistent) impact # 
*Careful! These poor excuses for academic publications often pick names that are very similar to the names of reputable journals!

The bogus academic journal racket is now officially out of control


10) Evidence-based vs. science-based approaches (in general, prefer the latter to the former)




Signs of bad science 
 

 

Oh, the irony...  
Sometimes pseudoscientists accuse "the establishment" of practicing bad science.

 

Discussion Q: List as many problems with the above image as you can.


Discussion: Homework

Pseudoscience and scientific reasoning 1


1. (a) Pick one supplement/herbal remedy/alternative medicine treatment that you or one of your family members uses or one that you are curious about.

(b) Find a website or article that promotes that treatment and read what their supporting evidence is. 
Then go to quackwatch.com and/or http://www.sciencebasedmedicine.org/ and search for the treatment. 
If your supplement/treatment doesn't come up on one of those sites go to google and type in the name and the work "debunked." 
Read the article you found.
(i) What does their interpretation of the evidence suggest? 

(ii) In light of the concepts we've learned today and throughout the class, write a short 1/2 page summary of your findings. 
Note: try not to focus too much on the issue of biases. 
Instead, focus on comparing the quality of evidence and arguments. Use the Rough Guide to Spotting Bad Science.
 

My pick: ThunderShirts for anxiety reduction in companion animals (esp. dogs). www.thundershirt.com

This is totes the calmest I've ever been

2. How to Make a Fad Diet. Make your own fad diet by following this handy-dandy guide. (You can skip Step 5). 

#1 – You need a catchy title, usually taking the form of “The blank Diet.” You can fill in the blank with almost anything. 

 

#2 – Make outrageous claims of success. The bigger the lie, the more people are inclined to think that it’s not a lie because no one would be that audacious. So just come up with a very impressive figure – a pound a day, 10 pounds a week, or whatever.

#3 – Testimonials. Personal stories, starting with your own, are the bedrock of fad diets. Don’t worry if there is absolutely no scientific evidence to support your claims – fad diets are not about evidence.

#4 – The Secret. Your fad diet has to have the secret or key to weight loss. Make this as compelling as possible, using language like, “unlocking the secret,” “hacking the body,” unleashing the genetic code,” or whatever.

(Optional) #5 – Cherry pick the research. If you want to add a little optional bling to your website, you can link to research supporting your diet. Now, there won’t be any research that actually indicates your diet works, but don’t worry about that. You can cherry pick studies that appear to support your diet.

#6 – Sciencey Explanations. This part can be real fun – make up BS explanations for why your diet works. There are some good old standbys if you can’t think of anything original, so don’t worry. You can’t go wrong with toxins. Just mention toxins a lot, and say that your diet eliminates toxins.

#7 – Demonize a food group. Every narrative needs a villain, and every fad diet needs a demonized food or food group. These are evil foods that make your body store fad and sap you of energy. Avoiding them is a key to the diet, and one of the secrets you discovered.

#8 – Attack your critics. You may get some blowback from scientists and doctors about how pseudoscientific and unhealthy your diet is, and be criticized because the research does not support your claims. Don’t worry about this… If you are feeling the pressure, however, then just attack your critics.






Gems from the class
(There are a lot of these, but they're worth a read. I was cracking up at your answers!) 

"This diet hacks into your brain and tricks itself to erase all your fat! The chemicals and diet work like a dynamic duo to round up all of your fat cells and scientifically expel them from your anus!"

"Exercise and eating healthy were never my thing, so when I heard about the raisin diet I was intrigued. I mean, raisins come from grapes and grapes make wine, so it can’t be that bad...
It’s not even hard at all! All you have to do is ONLY eat raisins for ten days and the pounds just fall off!"

"Scientists said I was crazy to put worms into my body, but now I am 200lbs lighter and never have been healthier!" 

"The secret of the fueled-up diet is to have the user drink nothing but kerosene and that’s it. Just drinking kerosene and not eating.
Kerosene is fuel and your body needs fuel to go and the more fuel you have in your body the more fat you will burn. Therefore, drinking kerosene will hyperburn fat cells at a high rate."

"[Our diet products contain] natural ingredients from the leaves of tropical plants located in the Caribbean Islands!... The leaves from the plants have special ingredients, ones that are activated through exercise and good eating habits."

"While on The Cat Diet try to avoid foods that dogs would eat. Foods such as beef and pork will sabotage the diet and should be avoided at all costs. For proteins, it’s best to stick with fish and other aquatic meats. 
... We surveyed skeptics of our diet and it shows that 99% of people who don’t believe in our diet also don’t like cats. Therefore, they can’t be trusted."

"Listening to this DVD helps breaks down toxins by becoming a toxin then earning the trust of the other toxins so that the toxin can kill the other toxins. Then once all the other toxins are gone the toxin kills itself by infecting itself with a toxin."

"Fruit roll ups have a very important chemical called tastyaf that create a shield around your stomach that prevent toxins from entering your bloodstream." 

"The Pancake and Beer Diet: The secret is in the way carbs unlock the real power of the body."

"Don’t listen to the naysayers! Our product is completely safe!* Science even says so! Would science lie to you? Also, results!" 
*Tape-Pellet™ is safe until you die from malnutrition.

Please don't swallow a tapeworm to lose weight...


 "Introducing The Hyper Drive Diet ™. Lose up to 5 pounds a week by eating science fiction novels. The fiber in the paper cleanses your body of toxins and helps to shed pounds fast!... The science in the books is absorbed into your blood, and neutralizes toxic chemicals!"
 




Another fun idea: Google "[Insert any food] diet" and skim the first page of results. Have you hit on a miraculous recipe for weight loss? (Remember our "[Insert any food] causes cancer" Google experiment?) 




Pseudoscience: An added complication 
Fake news and the Interwebs

As some of your classmates aptly noted in their homework responses, the Internet (fake info gains a foothold via social media, in particular) facilitates the spread of pseudoscientific B.S.

Critical thinking in the real world, or
Why this class is valuable:

Study Finds Students Have Dismaying Inability To Tell Fake News From Real 

Top 20 fake news stories "out-shared" top 20 real news stories about the election on social media 

(related: check out some examples of fake stories shared extensively on Facebook here)




Discussion Q

Have you or a friend come across a fake news story on social media recently? What was it? How did you figure out that the story was fake?





My example ...

Which people shared enough for it to end up on snopes.com 

A rough but handy guide to distinguishing real vs. fake news stories on social media:
  
How to recognize a fake news story 


And remember...

Snopes.com is your friend 

Evaluating scientific findings, continued*
*Also applicable to evaluating claims about scientific findings (e.g., news headlines) 

From Monday 11/28: Consider phase of testing, strength of evidence, and scope of findings when critically evaluating the efficacy of medical treatments.

Discussion: Homework

Pseudoscience #2


2. (a) Very often on social media and in the news you'll see reports or claims that "X has been found to cure cancer." Explain some of the things you'd need to know in order to properly evaluate this claim.

(b) Often, along with the claims that "X cures cancer" you'll read that the government or Big Pharma is suppressing it/making it illegal. Suggest some alternative explanation for why a government or drug research might not produce or continue research on a drug/plant, etc... 

More food for thought: 
 
1. Here's an awesome example of pseudoscience presented as credible science on the Internets:

Alzheimers caused by mercury in vaccines!!1! 
(Be very, very, very skeptical of what you read on this website)

 
How many problems can you find in this article?
Refer to the criteria listed above (Elements of a good clinical trial, what to look for when evaluating studies, a rough guide to spotting bad science).


2. Peruse this site and compare headlines vs. what the scientific findings actually support.
 

 

Wednesday 11/16: Exam #2 Review

Reminders:

• Regular office hours Wednesday 11/16
• Practice test here
• Review session Thursday 11/17, 6-8 pm in Shatzel Hall 317 (Liz and John's students welcome to attend)
• Please come prepared with questions!
• Exam #2 Friday 11/18

Today: Exam #2 Review  

Topics covered since Exam #1  

• Informal fallacies
• Ad hominem (& reverse ad hominem)
• Circumstantial ad hominem (& reverse)
• Genetic fallacy (& reverse)
• Tu quoque
• Poisoning the well
• Naturalistic fallacy types 1 & 2
• Argument from ancient authority
• Ad populum (bandwagon fallacy)
• Appeal to emotions
• Ad baculum (appeal to force)
• Appeal to unqualified authority

• Obstacles to good reasoning
• Vagueness
• Semantic Ambiguity
• Syntactic Ambiguity
• Group Ambiguity
• Fallacy of Equivocation
• Fallacy of Composition
• Fallacy of Division

• Obstacles to good reasoning, continued
• Weasel Words
• Euphemism/dysphemism 
• Misleading Comparisons 
• Omitted/ambiguous comparison class
• Apples-to-oranges
• Puffery

• Generalizations
• Structure of arguments to general conclusions
• Target and sample populations
• Sample size and representativeness
• Hasty generalizations
• Anecdotal evidence
• Biased samples
• Operationalizing terms 
• Margin of error 

• Statistical syllogisms
• Structure of a statistical syllogism (generalization inverted)
• Considerations:
• How common is the trait in the target population?
• Homogeneity of target population?
• Relevant sub-groups? 
• Manipulating your audience with numbers: Averages/percentages vs. absolute numbers

• Polling
• Structure of a polling argument (see generalizations)
• Selection bias
• Self-selection
• Measurement errors
• Common sources of measurement error:
1.  Medium
2. Vagueness
3. Timing
4. Place
5. Second-hand reporting
6. People are dumb but don't want to look it
7. Phrasing
8. Self-selection

• Causal reasoning
• Mill's Methods:
• Agreement
• Difference
• Joint Method of Agreement and Difference
• Method of Concomitant Variation
• Dose-response relationship
• Correlation vs. causation
• General structure of a causal argument 

•  Common errors in causal reasoning
• Post hoc ergo propter hoc
• Misidentifying relevant causal factors
• Mishandling multiple factors
• Cum hoc ergo propter hoc
• Wrong direction of causation
• No control group
• (Related: biases, conspiracy theories)

• Arguments from ignorance
• Structure
• Evaluating premises
• Fallacious arguments from ignorance
• Legitimate arguments from ignorance
• Argument by elimination

• Arguments from personal incredulity
• Structure
• Evaluating premises

• Anomaly hunting [week of 11/14?]

I know. That's a long list.

  
But don't despair!

Use the practice test as your guide as to what you should focus on
*Make sure to do the practice test!

Other pieces of advice:
-Obviously, review the lecture notes & this blog. But don't spend all your time reading!
-Try to do at least couple of the questions from each homework set without your notes, then check your answers. This will help you isolate your strengths and weaknesses.
-Check your old, completed homework assignments against the answer keys posted on Canvas. Make a note of answers that differ from yours especially if you don't understand why the answers differ. Make sure to ask about these questions during class, office hours, or the review session.
-Same goes for the quizzes we've taken since the last exam.
-Make a note of questions you don't understand from the practice test, and bring those questions to class, office hours, or the review session.
-Ask your peers for help. Form study groups. "Grade" each other's homework.
-Finally...

Don't panic

Friday 11/11:

I recommend that you start studying for Exam #2 now. 
In the course of your review... 
Please give some thought to what you'd like to review in our discussion on Wednesday, 11/16. Then, leave your suggestions in a comment below. 
Make sure to include your section time (9:30 or 11:30) in your comment.

I will update this post prior to our 11/16 discussion class.

Update 11/15 p.m.:

I haven't received any requests in the comments regarding what you'd like to cover tomorrow. Feel free to leave comments later tonight. In the absence of student feedback, we will cover:

-Questions about the practice exam
-Lingering questions about the homework
-Questions regarding the "identify the fallacy" examples Ami posted earlier this week
-A selection of questions from this 15 page set of practice questions I've created for you



In the absence of student feedback, I will focus mainly on generalizations, statistical syllogisms, polling, and causation.

I also want to review the answers to two questions from Quiz #5 that gave people trouble:

4. Most philosophers like cats. Liz and John are philosophers. Therefore, Liz and John probably like cats.

Name the argument structure: 

a) Generalization

b) Modus ponens

c) General causal claim

d) Bob

e) Statistical syllogism

6. Scientist testing new pain medication: "A dose of 50 mg causes the most pain relief and the fewest side effects. Below 50 mg, the rats still show signs of pain. Above 50 mg, the rats start getting sick." 

The above argument employs which of Mill's methods to reach its conclusion? 

a) Method of agreement
b) Joint Method of Agreement and Difference
c) Method of Concomitant Variation
d) Method of Difference
e) Method Man

Happy studying :)

 

11/9: Causal reasoning errors, arguments from ignorance and incredulity

 Announcements:

• Quiz posted on Canvas this afternoon (Weds 11/9), due by midnight!
• Exam #2 next Friday (11/18)
• Liz's review session: Thursday 11/7, 6-8 pm, Shatzel Hall 317. Remember, I'm also available during office hours.
• John's review session: Next week, probably Tuesday 11/5, John will provide details soon.
• Be sure to let me know in advance if you are unable to take the exam at the scheduled time. Makeups require advance notice. Serious & unavoidable conflicts only!
• No class Friday (11/11)!

  

 Plan for today:

• New material: arguments from personal incredulity
• This is first because I've got to cover it today
• Review arguments from ignorance
• Structure of argument & how to evaluate
• Legitimate vs. illegitimate (fallacious) arguments from ignorance
• Related concepts: falsifiability, burden of proof, null hypothesis, target audience, "proof," relevant experts
• Hopefully clear up confusion...
• Review HW 4.5
• IF TIME, Review HW 4.4 (Causal reasoning errors)
• Please see me during office hours to discuss questions we don't get to today! 

Arguments from personal incredulity

Incredulity = def. The state of being unwilling or unable to believe something

  
 
Note: An argument from personal incredulity is a type of argument from ignorance. However, arguments from personal incredulity rely on a premise that states "I don't know," rather than a premise that states "No one knows."

Structure of an argument from personal incredulity

P1) I find X implausible.
C) Therefore, X is false or implausible.

If it isn't already obvious what's wrong with this argument, add the implied premise to make it valid. 


What's the implied premise (enthymeme)? 
Hint: modus ponens it.
Why is the implied premise problematic? 


The problem with arguments from personal incredulity:

The fact that you don't believe X, you're ignorant of X, or you don't understand X is usually not a good reason to conclude that X is false, or highly improbable.

New term - epistemic humility
Basically means realizing your knowledge is limited
E.g., Socrates is wise because he knows how much he doesn't know


Examples of arguments from personal incredulity

• I don’t understand how life can just move from bacteria to humans all on its own. There had to be an intelligent designer behind it! [Subtext: evolution must be false!]

 

• I don't believe that .999... = 1, so it can't be true!

Regarding the second example, mathematicians will tell you this is true. I still can't wrap my brain around it, but here's a proof (via iflscience.com):

More crazy math facts here 

As these examples illustrate, an argument from personal incredulity usually indicates lack of knowledge, lack of effort to uncover information, lack of imagination, bias, etc. on the part of the arguer.

It's safe to assume such arguments are fallacious whenever the personal view expressed is contrary to that of experts.
 

Example from Homework 4.5:

7. I simply can't believe the God would allow the earth to overheat, therefore climate change must be false. (Paraphrasing Senator James Inhoff).

In class exercise:

a) Put this argument in standard "argument from personal incredulity" form. Include the enthymeme as P2!
b) Assess the premises (acceptable, unacceptable, questionable).
c) Give your overall assessment of the argument.
d) *Extra credit/for funsies - Formalize Inhoff's argument in support of P1, i.e., his disbelief in global warming. 
Hint: make it a valid, deductive argument (modus ponens). Then, assess this sub-argument's premises and its overall strength. 

If you're feeling confused at this point, here's a short video that explains arguments from personal incredulity quite nicely (though it's a bit boring):

Review: Arguments from ignorance

Basic structure

Formulation #1:
X is true because no one's proved that X is false

P1) No one's proved that X is false
C) Therefore, X is true





Formulation #2:
X is false because no one's proved that X is true

P1) No one's proved that X is true
C) Therefore, X is false 


[Fallacy ref confirms this formulation has the same problem]



A word about proof: Many arguments from ignorance rely on the assumption that the standard for proof is "beyond a shadow of a doubt." But this is an unreasonable standard of proof. The arguments we're considering are inductive and thus, the conclusion is always probabilistic. Accordingly, the proper standard of proof is something like, "Our best theories, evidence, and expert consensus support X conclusion." When I use the term "proof" it should be taken to imply "proof" in the latter sense.


Our formal structure (BGSU PHIL 1030 structure) for arguments from ignorance is modified to indicate that the argument is inductive, i.e., probabilistic, and that evidence in support of the conclusion should come from the right kinds of sources (relevant experts). We reconstruct arguments from ignorance as two premise arguments.


Formulation #1:
X is true because no one's proved that X is false

P1) There's no proof (i.e., no good evidence) that X is false
P2) There's been a reasonable search for the relevant evidence by whoever is qualified to do so (often experts)
C) Therefore, X is probably true


Formulation #2:
X is false because no one's proved that X is true

P1) There's no proof (i.e., no good evidence) that X is true
P2) There's been a reasonable search for the relevant evidence by whoever is qualified to do so (often experts)
C) Therefore, X is probably false



Arguments from ignorance are often, but not always fallacious (bad) arguments.


Inductive, scientific arguments often take the form of a legitimate (cogent) argument from ignorance:

P1) No one's proven (i.e. there's no good evidence) that vaccines cause autism.
P2) There's been a reasonable search by the relevant experts for evidence that vaccines cause autism.
C) Therefore, vaccines probably don't cause autism.

 Notice that in this example the argument goes through because we commonly accept that the person who wishes to challenge scientific consensus bears the burden of proof. That is, it's on the challenger to provide positive evidence that can change our minds.

 

How to differentiate legitimate, i.e., cogent, from illegitimate, i.e., weak, arguments from ignorance: 

• Assess P1 and P2 to see if they are acceptable, unacceptable, or questionable.  
• If P1 and/or P2 is/are unacceptable (or highly questionable), the argument is fallacious.

General guidelines:

• Generally, if an argument from ignorance takes the following form it's bad:
  
  There's no evidence that X doesn't exist, therefore it does.
  
  Or (same thing, different wording):
  
  You can't prove that X doesn't exist, therefore it does.

Explanation: The arguer is making a non-falsifiable claim. You can't prove something exists with the absence of evidence; otherwise we could prove absolutely anything exists.

• Generally, the burden of proof falls on the person making an existential claim. 
• That is, if you want to claim that something exists or that some variable explains an observation, the burden is upon you to provide evidence for it.
• Otherwise, the reasonable position is the null hypothesis. The null hypothesis just means that we assume no entity or phenomena exists unless there is positive evidence for its existence. 

• Generally, the burden of proof in an argument rests on a person who challenges expert/theoretical consensus (or, in some cases, the consensus common knowledge supports). 

 

• A note regarding evidence: The evidence should be something the target audience can accept. A safe way to proceed is to assume that you're trying to persuade a universal audience. 

 

• Keep in mind that there are different kinds of evidence
• Direct evidence (roughly, what we've been calling positive evidence): the evidence directly supports a conclusion, no inference required
• I see Professor Plum kill Mr. Body
• Indirect evidence (in law this is called circumstantial evidence): the available evidence supports an inference to the conclusion
• Professor Plum had motive, he was the only one present when Mr. Body died, and he's got blood all over his clothes

(paraphrased from this source)

Both direct and indirect evidence can provide strong support for a conclusion. 

On that note, consider an example from the homework:


3. There's no evidence of life in the universe besides on earth, therefore there probably isn't any life elsewhere in the universe.  

Groups:

a) Put the argument in standard "argument from ignorance" form
b) Assess the premises for acceptability
c) Provide an overall assessment of the argument's strength 

 

                       
Let's all take a minute to ponder the fact that someone knitted this for his/her cat.


Homework 4.5 Review, continued
(questions not covered above)

a) Put the argument in standard "argument from ignorance" form
b) Assess the premises for acceptability
c) Provide an overall assessment of the argument's strength 


1. Aliens are visiting us since no one has proven that they aren't.

2. There's no good evidence to believe aliens are visiting us, therefore they haven't.

4. You can't prove that CERN particle collider won't cause a tear in the space time fabric leading to the end of the universe, therefore it will.

5. There's evidence you don't understand arguments from ignorance, therefore you can understand how to evaluate this question.

6. No one has ever disproven Bigfoot's existence, therefore, he probably exists.



Lesson 4.4 (Review)
1) General structure of a causal argument
2) Common errors in causal reasoning


1) General structure of a causal argument:

P1) X is correlated with Y.
P2) The correlation between X and Y is not due to chance (i.e., it is not merely statistical or temporal).
P3) The correlation between X and Y is not due to some mutual cause Z or some other cause.
P4) Y is not the cause of X. (Direction of causation.)
C): X causes Y. 
*In P2, to show the correlation between X and Y is not due to chance, there should be a (plausible) proposed causal mechanism.

  
2) Common errors in causal reasoning

A causal argument or explanation is strong to the degree that we are willing to accept each of the four premises. If we find one or more of the errors listed below we should be less confident about the affected premise(s). 

1. Post hoc ergo propter hoc (after therefore because of)--usually referred to just as "post hoc fallacy"). 
-Confusing causation with temporal order. 
-Just because Y happened after X it doesn't follow that X caused Y. -This error applies to (P1), (P2), and (P3).

"After a black cat crossed my path, I failed my critical thinking exam. That cat caused me to fail!"

U failed lol

 

2. Misidentifying the Relevant Causal Factor(s)
-There are often hundreds of factors common to each causal event. It does not follow that they are all relevant.
-Misidentifying the relevant causal factor is usually a consequence of not having enough knowledge of the topic at hand. 
-This error applies to (P1), (P2), and (P3). 




Clueless pet owner #1: "I know she's a bit heavy... I read that her breed puts on weight easily."

Veterinarian: "Yes, beagles do put on weight rather easily. But genetics are not the main reason your dog resembles a bloated coffee table. The relevant causal factor is how much you're (over)feeding her."


3. Mishandling Multiple Factors: 
-For every general causal argument there will often be many antecedent variables involved. 
-Identifying the one that has causal import can be tricky. 
-People will often fail to consider alternative - and perhaps more plausible - causal variables to the one(s) they identify. 
-This error applies to (P1), (P2), and (P3).

Clueless pet owner #2: "My cat is really itchy. I think she has a grass allergy/food allergy/sensitive skin because she's a Scorpio..."


Veterinarian: "Is she on flea prevention?"


C.O. #2: "No way. That stuff is full of chemicals!"


Veterinarian: "Yeah..."

4. Confusing Correlation and Causation (Cum hoc ergo propter hoc="comes with therefore because of")
-Just because two events or variables are correlated or co-occur, it doesn't follow necessarily that there's a causal relationship. 
-Often a good way to avoid committing this error is to see if you can come up with a likely causal mechanism. If you can't then it's likely simply correlation.
-This error applies to (P2) and (P3).





 
5. Confusing Cause and Effect (aka Direction of Causation). -Often it is difficult to disentangle the direction of causation. Does X cause Y, or does Y cause X?
-This error applies to (P4). 

E.g., "Alcohol abuse causes depression."

What's the alternative direction of causation? Is it equally plausible?


Less plausible:

6. No Control (see Mill's Methods: Method of Difference). 
-Often misattributions of causation occur because there is no control group. 
-E.g., If we don't know the natural prevalence rate of a disease or its average natural healing time we cannot reasonably attribute causal power to a purported remedy. 
-Applying a control helps eliminate errors in (P1), (P2), (P3), and (P4).

Specific to medical/health-related research: Need to control for the placebo effect.

 



Lesson 4.4 - Review selected homework questions

I. (a) Identify the proposed cause and effect. (b) Suggest alternative explanations. (c) Suggest which premise of a standard form causal argument the original explanation fails and name the error.

Example:

Frank reads a lot and wears glasses therefore reading a lot must cause permanent eyesight damage.

(a) Reading a lot caused Frank to lose his eyesight.
(b) Genetic factors might also explain why Frank's eyesight is poor.
(c) This argument would fail Premise 2 because it commits confusing causation with correlation (cum hoc ergo proptor hoc).

 
1. Mary is feeling sick. She ate about 2 hours ago, it must be something she ate. 

2. Did you hear about the pipeline rupture last week? I'll bet those anti-pipeline protestors sabotaged it in order to make everyone afraid of pipelines. 

3. I feel hungry. That explains why I feel so tired. 

5. He got his well-paying job because he dresses well.

6. Violent video games cause children to be violent.

7. You shouldn't smoke the pot. My friend smoked the pot and now he has a mood disorder.

9. Bob sexually harassed his co-worker because he watches pornography. 




III. Critical Thinking in the Real World

As part of the war on drugs, Black and Hispanic citizens (especially men) are subject to stop and frisk at disproportionately higher rates than white citizens. Black and Hispanic drivers are also pulled over at disproportionately higher rates as part of the war on drugs.

Police and others explain their behavior by pointing to incarceration rates. That is, the fact that more Blacks are arrested and convicted for drug crimes causes police officers to scrutinize them more. For every white male in prison there are 6 Black males, about 60% of which are there for drug crimes (mostly minor). It turns out, however, that Blacks and whites use and sell drugs at approximately the same rate (within a percent). Think about direction of causation and feedback loops to explain why it is both true that Blacks and hispanics are arrested and convicted of drug crimes at higher rates that whites even though there's no difference is rates use or selling.