LSAT - Logical Reasoning Bible
LR Primary Objective #5
Carefully read and identify the question stem (Question Type). Do not assume that certain words are automatically associated with certain question types.
(Formal) "Reversibility" (def.)
Certain relationships between 2 vars have exactly the _same_ meaning regardless of which "side" of the relationship is the starting point of your analysis (SOME is a classic example) (E.g. A <-s-> B can go both ways; is reversible)
(PR) Take any principle and you can...
Convert it into a conditional statement! (i.e. Sufficient --> Necessary) (But DON'T remove the judgment bit!)
(Weaken) If you see a gap or hole in the argument...
Immediately consider that the correct answer might attack this hole
(Assumption) If you see a conditional conclusion and then are asked an Assumption question...
Immediately look for an answer that confirms that the Necessary condition is _truly necessary_ or that eliminates possible alternatives to the necessary condition (Sufficient --> Necessary)
(Causality) When you identify a basic causal relationship in the _conclusion_ of an LSAT problem...
Immediately prepare to either _weaken_ or _strengthen_ the argument
(Causality) Differences Between Causality and Conditionality
In a conditional relationship... 1.) The chronology of the 2 events can differ (unlike a causal relationship where one event must precede the other) 2.) Often related directly (i.e. one _makes_ the other happen) BUT don't have to be
(Common Flaws) Relativity Flaw
Information about a relative relationship -- one involving comparison -- is used to draw an absolute conclusion, or when a relative conclusion is drawn from absolute info (e.g. "Valentina is the tallest in the class, therefore she is _tall_" or "Valentina is tall, therefore Valentina is _the tallest_ in the class)
If you want to Weaken an argument...
Isolate the conclusion!
(PR) Whenever there is an element of judgment in there...
It's a Principle (PR) question [overlay] (e.g. a 'should')
(Assumption) If the question stem is word salad...
Just logically negate ALL the answer choices and see which deflates arg (don't waste too much time, if you eliminate off the bat the word salad will bite you)
(MoR) You should carefully study all MoR Answers -- correct and incorrect:
Keep a list of all the different types of reasoning methods you encounter! (They tend to get rehashed)
(Justify) As you examine each question to apply the Mechanistic Approach...
Keep in mind that even if an element initially appears 'new' or 'rogue', it could still be elsewhere but in a _reworded_ form.
Argument Part (AP) questions are a subset of...
Method of Reasoning Questions
(Parallel) If the question stem does NOT mention flawed reasoning...
the stimulus could still contain flawed reasoning! (Never relax on the LSAT!)
In a stimulus with complex terminology...
the test makers often prey upon that difficulty by presenting answers that have a degree of truth but go a bit too far. (Watch your modifiers!)
(Assumption) The consequence of negating an assumption is that...
the validity of the conclusion is called into question
(Causality) If a causal claim is made in the premises...
then _usually_ no CAUSAL reasoning error exists in the argument (though of course it might be flawed in other ways)
(Assumption) If the argument gets blown all to bits by logically negating an answer choice...
then bam, you know that assumption was necessary. It's the correct answer.
(Flaw) Arguments that draw causal conclusions are often flawed because...
there may be another explanation for the stated relationship
Probability Indicators
Must Will Always Not always Probably Likely Would Not necessarily Could Rarely Never
(Formal) Does an inference have to be makeable from both sides in order to be valid?
NO (some only work when addressing the chain from one direction)
In an MoR question, there is absolutely...
NO NEED to... analyze the _strength_ or _validity_ of an arg
(Formal) At the lowest rung of the Logical Ladder (SOME)...
NO _inherent_ inferences follow
(Formal) Is there a contrapositive for SOME and MOST statements?
NO! (only things like ALL have contras)
(Assumption) The Better Negation Test
Negate the argument WHILE TRYING TO MAKE IT WORK, helping you to more easily pick out the one that truly wrecks the arg when negated (see https://www.reddit.com/r/LSAT/comments/3ck93a/how_to_negate_the_full_version/)
(Assumption) Logical Opposite of SOMETIMES
Never
The most tempting wrong answer choice sometimes appears just before the correct answer...
Never relax during the LSAT!
(Assumption) Logical Opposite of SOME
None
(Formal) The Negative Logic Ladder
None | | v Most are not | | v Some are not
(Formal) The Reversible Relationships (table)
None (<-|->) Some (<-s->) Double-arrow (<-->)
(Assumption) Logical Opposite of ALL
Not All
(Assumption) Logical Opposite of ALWAYS
Not Always
(Assumption) Logical Opposite of EVERYWHERE
Not Everywhere
Note that the premises and conclusions can be presented in any order!
Note that the premises and conclusions can be presented in any order!
(Assumption) Logical Opposite of SOMEWHERE
Nowhere
(Formal) Additive Inf. Principle #11
Once an inference bridge is built, it does NOT need to be built again
Must Be True (Q Indicators)
One or both of: 1.) The stem often indicates the info in the stimulus should be taken as true (e.g. if statements above are true...) 2.) The stem asks you to ident a single answer choice that is proven or supported (Stem indicates that one of the answer choices is proven by info in the stimulus)
(Eval) If the answer choice is correct...
One response will WEAKEN the arg, and the other will STRENGTHEN it
(Formal Logic) When you diagram statements involving SOME...
Place a double-arrow with the letter "S" between the 2 elements (e.g. X <-s-> Y for 'some Xs are Ys')
(Formal) When you diagram a statement involving Most or 'most are not'...
Place an M over a single-headed arrow (e.g. X -m-> Y for 'most Xs are Ys') (W -m-> /Z for 'most Ws are NOT Z's)
(Assumption) The Logical Opposite is NOT the same as the...
Polar Opposite
(Justify the Concl.) The Justify Formula
Premises + Answer Choice == Conclusion
LR Primary Objective #6
Prephrase: after reading the question stem, take a moment to mentally formulate your answer to the question stem.
(Assumption) 'Defender' Assumption
Protect the arg by eliminating ideas that could weaken it
Remember...
Read CLOSELY and pay strict attention to the MODIFIERS used by the author.
LR Primary Objective #4
Read closely and know precisely what the author said. Do not generalize!
LR Primary Objective #2
If the stimulus contains an argument, identify the conclusion of the argument. If the stimulus contains a fact set, examine each fact.
If you need to, apply the why-because test to an answer choice.
If you can't put it in the [conclusion] box, then it's not a possible answer.
(MoR) Skip over those answer choices with 'spinning swords' in an attempt to find a simpler one that makes more sense
Revisit the tangled ones later if you come up tails (saves you much time)
(Formal) Remember, going backwards against a [directional] arrow is the inherent inference ______ ....
SOME, and thus we can ride the Some Train
The understanding required for Must Be True:
What did you read in the stimulus, and what do you know on the basis of that reading?
To identify conclusions, ask yourself...
What is the author driving at? What does the author want me to believe? What point follows from the others?
To identify premises, ask yourself...
What reasons has the author used to persuade me? Why should I believe this argument?
(Resolve) Because you are not seeking to disprove one side of the situation, you must select the answer choice that contains...
a POSSIBLE CAUSE of the situation
(# & %) Just because a problem mentions a number or percentage...
does NOT necessarily mean that the problem uses that information centrally within the argument
When determining specific conditional logic relationships...
sometimes it's easier to get the contrapositive first, and back out the original statement
(Flaw) Flaw in the Reasoning questions are exactly the same as MoR questions except...
the question stem indicates that the reasoning in the stimulus is FLAWED
Most Strongly Supported
What can we [all but] PROVE?
"usually" can translate to...
"Most"
Assumption 'Flaw' Question Stems (arg is based on a flawed assumption)
"The argument is most vulnerable to the criticism that it takes for granted that..." "The argument is flawed because it takes for granted that..."
(Assumption) Logical Opposite of "to be rich, you must be smart"
"To be rich, you do NOT NECESSARILY have to be smart"
When the phrase "then either" is used...
"either" translates to "both" (but individually, not combined) (e.g. "Desmond likes Biology better than either Chemistry or Physics" == "Desmond likes Biology better than BOTH Chem. or Phys. (indiv.))
Logically, "tends" means...
"most[ly]"
"If and only if" synonyms list
(If get something more exciting than these, just hash it out logically via separate statements and link) If but only if Then and only then Then but only then When and only when When but only when All but only
# & % Misconceptions
- Misconception #1: Increasing/decreasing _percentages_ automatically lead to increasing/decreasing _numbers_ (i.e. DON'T assume higher % == higher # !!) - Misconception #2: Increasing/decreasing _numbers_ automatically lead to increasing/decreasing _percentages_ -Misconception #3: Large numbers == large percentages - Misconception #4: Large percentages == large numbers
(Point at Issue) You can double-check your answer via the...
Agree/Disagree Test: The correct answer MUST produce responses where one speaker would say "I totally agree" and the other speaker would say "No way that's BS, man"
The correct answer to a Must Be True question...
...can always be _proven_ by referring to the facts as stated in the stimulus. (CERTAIN to occur. Watch out for answers the are _likely_ or _possible_)
Make sure to carefully reach each stem, and...
...do NOT assume that a particular word always indicates a certain question type.
The information needed to answer the question always resides...
...in the stimulus, either implicitly or explicitly.
If you are unsure about which LR diagramming approach to take...
...look to diagram LESS rather than more. (though it will be inevitable for certain crazy problems)
Complex arguments contain...
...more than 1 conclusion (A main conclusion and other subsidiary conclusions)
Even seemingly broad speaker idents can tell you...
...more than you'd expect, including potentially valuable info about the topic to come or possible speaker biases (can tip you off to what about to read, or even that a flawed argument is coming (eg. 'Advertisement' or 'Politician'))
Must Be True questions...
...require you to select an answer choice that is _proven_ by the information presented in the stimulus.
Reading an LSAT stimulus is about...
...seeing PAST the topic to analyze the structural relationships present in the stimulus.
A complex argument...
...takes an initial conclusion and then uses it as a premise for another conclusion (statement can be both a conclusion and a premise for another)
The strength of an argument is based solely on...
...the degree to which the premises prove the conclusion.
You can often predict the occurrence of Must Be True questions because...
...the stimulus of _most_ of these questions does NOT contain a solution.
(Causality) Many causal conclusions are flawed because
...there can be alternate explanations for the stated relationship
The counter-premise is NOT integral to the author's argument structure because...
...when a counter-premise appears, the author usually indicates that it's not relevant or is outweighed by other considerations.
Quantity Indicators
All Every Most Many Some Several Few Sole Only Not all None
(Formal) The Non-Reversible Relationships (table)
All (-->) Most (-m->).
(Formal) Numerical Formal Logic Table
All = 100 Most = 51 - 100 (a majority) (can include all!) Some are Not = 0 - 99 (also "Not All") Most are Not = 0 - 49 Some = 1 to 100 ("at _least_ one") None = 0
3 common 'Weaken' question logical fallacies:
1.) Incomplete Information (Author fails to consider all of the possibilities, or relies upon evidence that is incomplete) 2.) Improper Comparison (Author attempts to compare Apples to Oranges) 3.) Overly Broad Conclusion (Author draws a conclusion too broad or more expansive than premises support)
(Justify) Sometimes you get a Justify fill in the blank (FIB) question. 2 telltale signs that it's a Justify:
1.) The blank in the stimulus is typically preceded by a premise indicator (stuff like 'because' or 'since') 2.) The wording of the question stem indicates that you are expected to justify the conclusion
MoR Incorrect Answer Types
1.) "New" element answers (if it describes something that did not occur or describes a "new" element, it's NOT correct) 2.) Half Right, Half Wrong Answers 3.) Exaggerated Answers (Be careful though! Just b/c extreme, not necessarily incorrect) 4.) The Opposite Answer 5.) The Reverse Answer
(Formal) The 2 Rules of Diagram Creation
1.) Always combine common terms (linkage) 2.) There is no traditional direction in logic (i.e. you can infer forward or backward)
Resolve question stems typically contain these features
1.) An indication that the answer choices should be accepted as true 2.) Key words that indicate your task is to resolve a problem
2 typical types of INCORRECT answers in Main PT questions
1.) Answers that are true but do not encapsulate the author's point 2.) Answers that repeat premises of the argument
(Common Flaws) Appeal Fallacies
1.) Appeal to Authority (Uses opinion of an authority in attempt to persuade the reader, even though auth. might not have relevant knowledge or all the info) 2.) Appeal to Popular Opinion/Appeal to Numbers (states that position is true b/c majority believe it to be true) (the issue here == NO PREMISES to support conclusion) 3.) Appeal to Emotion (emotions or emotionally-charged lang. used to persuade reader)
(Common Flaws) Mistaken Cause and Effect
1.) Assuming a causal relationship on the basis of the sequence of events 2.) Assuming a causal relationship when only a correlation exists (Remember, correlation does NOT imply causation) 3.) Failure to consider an alternate cause for the effect, OR an alternate cause for both the cause AND the effect 4.) Failure to consider that the events may be reversed (author mistakes effect for cause)
INcorrect answers in Must Be True and Most Strongly Supported Questions
1.) Could Be True or Possibly True Answers (contrary to the idea of the question, as an ironclad[ish] statement is sought) 2.) Exaggerated Answers (stretch the scope of info in stimulus to make broader statement) 3.) "New" Information Answers (b/c bring new, irrelevant stuff in, increasing scope) (NOTE: still read carefully, as the 'new' info might just be a restatement or synthesis) 4.) The Shell Game (A very similar idea (to that in the stimulus) appears in answers, but changed _just enough_ to be incorrect but still attractive) (something incorrect is discreetly slipped in) 5.) Opposite Answer (provides an answer that is completely opposite of the stated facts of the stimulus) 6.) Reverse Answer (contains familiar elements from the stimulus, but rearranged and flipped around in a way that is not correct)
(Point at Issue) Incorrect Answers
1.) Ethical vs Factual Situations (i.e. factual answers do not capture the point at issue in ethical situations, and vice-versa) 2.) Dual Agreement or Dual Disagreement (i.e. an issue that BOTH speakers agree/disagree with is tangential & INcorrect) 3.) The View of One Speaker is Unknown
(Eval) I'm not fond of the Variance Test. Do this instead, makes more sense.
1.) Identify the Flaw (the 'gap' between the support and conclusion) 2.) Fill it with the answer choice that comes as close as possible to filling it
To effectively Strengthen an argument...
1.) Identify the conclusion -- this is what you are trying to strengthen! 2.) Personalize the argument 3.) Look for weaknesses in the argument (i.e. a weak spot is ripe for an answer choice that fills it, thus strengthening the arg) 4.) Args that contain analogies or use surveys rely upon the validity of those analogies and surveys (generally, choose an answer that strengthens analogy or validity) 5.) The correct answer can strengthen the arg just a little OR a lot
(Common Flaws) Errors in Assessing the Force of Evidence
1.) LACK of evidence for position is taken to prove position is FALSE 2.) LACK of evidence against a position is taken to prove position is TRUE 3.) SOME evidence against a position is taken to prove position is FALSE (i.e. evidence presented only _weakens_, doesn't _disprove_) 4.) SOME evidence for a position is taken to prove position is TRUE (i.e. _partial_ support does NOT make arg invincible)
MoR Attack Strategy
1.) Make a general, abstract prephrase of what occurred in the argument and 2.) Apply the Fact Test (If an answer choice describes an event that did NOT occur in the stimulus, then that answer is NOT correct (akin to MBT)) 2.) Rigorously examine each answer choice to see if the test makers have created an answer that paraphrases your prephrase
(# & %) If you see a number or percentage in a stimulus...
1.) Make sure that the # or % is the focus of the problem, and not merely a distractor element. 2.) If it's not a distractor, identify whether a # or % is bing discussed and determine whether those figures represent the _overall_ total or just a _portion_ 3.) Always look for what is missing 4.) Watch for movement in any #, %, or total, and make note of whether it is rising or falling 5.) If a conclusion is present, often the conclusion is faulty and based on a confusion between the # and % idea
(Parallel) The 4 Tests to evaluate answers (in ~order of usefulness)
1.) Match the Method of Reasoning 2.) Match the Conclusion 3.) Match the Premises 4.) Match the Validity of the Argument (You can stop this train early if you successfully eliminate 4 answer choices early on in the steps) (If you get stuck on a step, move on to the next)
(Common Flaws) Errors of Conditional Reasoning
1.) Mistaking a NECESSARY condition for a SUFFICIENT condition 2.) Mistaking a SUFFICIENT condition for a NECESSARY condition (These are essentially Mistaken Negations & Mistaken Reversals) (remember: Sufficient --> Necessary)
The 13 Logical Reasoning Question Types
1.) Must Be True/Most Supported 2.) Main Point 3.) Point at Issue/Point of Agreement 4.) Assumption 5.) Justify the Conclusion 6.) Strengthen/Support 7.) Resolve the Paradox 8.) Weaken 9.) Method of Reasoning 10.) Flaw in the Reasoning 11.) Parallel Reasoning/Parallel Flaw 12.) Evaluate the Argument 13.) Cannot Be True
Assumption Question Workflow
1.) Narrow the field (peel away the obvious wrong answers) 2.) Logically negate the contenders 3.) The negated answer that WEAKENS the argument will be the CORRECT answer
(Resolve) Besides the discrepant or contradictory facts, most Resolve stimuli contain the following features
1.) No conclusion 2.) Language of Contradiction (but, however, etc.)
(Strengthen) Some incorrect answer traps
1.) Opposite Answers 2.) Shell Game Answers 3.) Out of Scope Answers
(Weaken) Common Incorrect Answers
1.) Opposite Answers (strengthen instead of weaken) 2.) Shell Game (idea/conclusion changed just enough to be incorrect but still attractive) 3.) Out of Scope Answers (miss the point and raise unrelated or tangential issues)
Two types of answers that will always be correct in Must Be True and Most Strongly Supported
1.) Paraphrased Answers (restate the stimulus in different terms) (can sometimes seem too obvious) 2.) Answers that are the sum of 2 or more stimulus statements (combo answers)
(Justify) The test makers have taken to creating questions that try to evade applications of the Mechanistic Approach, and their 2 primary avenues of deception are:
1.) Reworded Elements (e.g. the components stay the same, but a 'boycott' is reworded to an 'action', etc.) 2.) Distractor Elements (either background info or are not essential to the argument structure)
(Weaken, Causality) The most effective options to weaken a possible or probable causal conclusion are usually to...
1.) Show that the stated relationship is reversed 2.) Show that a statistical problem exists with the data used to make the causal statement 3.) Raise a possible third cause that could have caused both elements
(PR) The 2 Guiding Principles of Principle Questions [overlays]
1.) Stay within scope. 2.) Principles are just conditional rules.
Assumption question stems typically contain...
1.) Stem uses word "assumption", "presupposition", or some variation 2.) Stem NEVER uses the word IF or any other sufficient condition indicator (otherwise, it'd be a Justify or a Strengthen) (Will likely contain a necessary condition indicator such as 'required' or 'unless')
Most Justify question stems contain some or all of these 3 indicators or features:
1.) Stem uses word 'if' or another sufficient condition indicator 2.) Stem uses phrase "allows conclusion to be properly drawn" or "enables the conclusion to be properly drawn" 3.) Stem does not lessen the degree of justification (i.e. _not_ 'most justify' (Strengthen), we're looking for _ironclad_ justification)
Read the parts in the order given!
1.) Stimulus 2.) Question Stem 3.) EACH of the 5 Answer Choices DON'T read the Question Stem first!
(Eval) In order to determine the correct answer choice on an Eval question, apply the Variance Test
1.) Supply 2 POLAR opposite responses to the question posed _in the ANSWER choice_ 2.) Analyze how the varying responses affect the conclusion in the stimulus 3.) If different responses produce DIFFERENT effects on the conclusion, then the answer choice is CORRECT
(Common Flaws) Survey Errors (what invalidates survey results)
1.) Survey uses a biased sample 2.) The survey questions are improperly constructed (i.e. confusing or misleading) 3.) Respondents to the survey give inaccurate responses (e.g. 'what is your age?' or 'how much money do you make?')
(Parallel) These elements MUST be paralleled:
1.) The Method of Reasoning (If you recognize the form of reasoning used in the stimulus (e.g. causal, conditional, etc.), immediately attack the answers and search for the one with similar reasoning) 2.) The Validity of the Argument (i.e. if the conclusion is valid, eliminate the answers with invalid conclusions (and vice-versa)) 3.) The Conclusion (You MUST match the "certainty level" or "intent" of the conclusion (e.g. absolutes, logical force, etc.), but not necessarily the specific wording) 4.) The Premises (same wording levels as #3 Conclusion) (take this step AFTER checking conclusion)
Keep these fundamental rules in mind when approaching Strengthen, Justify the Conclusion, and Assumption questions...
1.) The stimulus will almost always contain an argument (ident, isolate, asses premises and conclusion) 2.) Focus on the conclusion (almost all correct answer choices impact it) 3.) The information in the stimulus is suspect (often reasoning errors present) 4.) These questions often yield strong _prephrases_ 5.) The answer choices are to be accepted as given, _even_ if they include "new" info
(Parallel) These elements do NOT need to be paralleled:
1.) Topic of the stimulus 2.) The order of presentation of the premises and conclusion in the stimulus (i.e. the order of presentation does not affect the logical relationship underneath)
"If and only if"'s effects (<-->) can also be achieved by...
1.) Using phrase "vice versa" 2.) Repeating and reversing the terms
3 Quirks of Assumption Question Answer Choices
1.) Watch for answers starting with 'at least one' or 'at least some' (a disproportionate number of these are _correct_) (still use the Assumption Negation Tech. !) 2.) Avoid answers that claim an idea was the most important consideration for the author (in _every_ Assumption question these answers have been _wrong_!) 3.) Watch for the use of "not" or negatives in assumption answer choices (Do NOT rule out a negative answer choice just because you are used to seeing assumptions as a positive part of the argument! It might still be correct)
Method of Reasoning questions feature the following information structure:
1.) You can use ONLY the information IN THE STIMULUS to prove the correct answer choice 2.) Any answer choice that describes an element or situation that does NOT OCCUR in the stimulus is NOT CORRECT
(Weaken) Overall Strategy
1.) Zero in on the conclusion 2.) Identify the most vulnerable hole (where the premises don't quite justify the conclusion) 3.) Prephrase! (using the vulnerability just discovered) 4.) Proceed to the answers (still need to elim, even if find a strong one!)
(MoR) Method--AP questions _often_ feature...
2 conclusions (a Main and a Subsidiary) (where the Main conclusion is typically placed in the 1st or 2nd sentence, and the Subsidiary typically in the _last_ sentence)
A "if and only if" B (diag.)
A <--> B (A DOUBLE-HEADED arrow!) (NOTE that if one fails to occur, they both fail to occur (contra. of the DH arrow))
(Formal) The correct way to reverse a "some are not" diagram (example)
A <-s-> /B As-is: Some As are not Bs REVERSED: Some things that are not Bs are As (i.e. some 'not-Bs' are As)
Counter-Premise v. Additional Premise
A Counter-Premise fills a potentially-damaging point of exposure, whereas an Additional Premise offers up something else to support the solution (for good measure) that is _sometimes_ non-essential to the conclusion.
(Causality) The classic mistaken cause and effect reasoning referred to often in LRB occurs when...
A causal assertion is made in the CONCLUSION, or the conclusion presumes a causal relationship (i.e. the author misattributes something to be the cause of another, and concludes as such)
Premise (def.)
A fact, proposition, or statement from which a conclusion is made. (the REASONS author gives to support conclusion)
Conclusion (def.)
A statement or judgment that follows from one or more reasons.
(Formal) The SOME train (cont'd)
A successful 'journey' (defined as a journey of at least 2 stops) yields an [additive] inference (start at the 'open' variable IN THE 'SOME' UNIT) (note that the SOME train can go right to left OR left to right (direction-agnostic))
(Weaken) Remember...
A wrong answer choice that weakens the passage but does NOT address the conclusion specifically is STILL WRONG!
(Causality) To attack a basic cause and effect relationship in weaken questions, will likely involve one of these tasks:
A.) Find an alternate cause for the stated effect B.) Show that even when the _cause_ occurs, the effect does NOT occur C.) Show that even though the _effect_ occurs, the cause did NOT occur (this works b/c in LSAT-world, the cause is the ONLY cause capable of bringing about this effect) D.) Show that the stated relationship is reversed/backwards E.) Show that a statistical problem exists with the data used to make the causal statement
(Formal Logic) MOST includes the possibility of...
ALL! (Flip side too: "most are not" can mean none)
"Either/or" LSAT meaning
AT LEAST ONE of the 2 (NOT EXCLUSIVE OF BOTH!!) (unless 'not both' specified)
(Point of Agreement) You can double-check your answer via the...
Agree/Agree Test (adaptation of Agree/Disagree): The correct answer choice MUST be one about which BOTH speakers would say, "Yes, I agree with that statement." If each speaker does not produce that response, the answer is INcorrect. (Use it sparingly, save time.)
In practice... (Weaken)
Almost all correct LSAT 'Weaken' question answers leave the premises untouched (would make for overly obvious answer choices)
To Strengthen a basic Causal conclusion, do the _exact opposite_ as for a Weaken question...
Almost always involves performing one of these tasks: A.) Eliminate any alternate causes for the effect B.) Show that when the cause occurs, the effect does indeed occur C.) Show that when the cause does NOT occur, the effect does NOT occur (i.e. show that the cause --> effect relationship is valid and strong) D.) Eliminate the possibility that the stated relationship is reversed (i.e. it HAS to work in the prescribed way) E.) Show that the data used to make the causal statement ARE accurate, or eliminate problems with the data
Always be aware of the discrepancy between the "LSAT world" and the "real world"!
Always be aware of the discrepancy between the "LSAT world" and the "real world"!
In Justify questions featuring conditionality...
Always be ready to ident and avoid Mistaken Reversals and Mistaken Negations
LR Primary Objective #7
Always read EACH of the 5 answer choices.
"Assumption" in a LR question means...
An _unstated_ premise, what MUST be true for the argument to be true/valid. (What is _taken for granted_ while making an argument)
(Common Flaws) Numbers and Percentage Errors
An author improperly equates a percentage with a definite quantity, or when an author uses quantity info to make a judgment about the percentage rep'd by that quantity (e.g. "increase in market share" (%) == "increase in overall revenue" (def. qty.))
(Common Flaws) Self-Contradiction
An author makes conflicting statements
(Common Flaws) Straw Man
An intentionally misrepresented proposition is set up because it is easier to defeat than the real argument (often accompanied by "what you're saying is" or "if I understand you correctly")
(Formal) Additive Inf. Principle #10
Analyzing Compound Statements When working with compound statements (statements with >=4 vars): A.) Recycle your inferences to see if they can be used to create further inferences B.) Make sure to check the closed vars (Kind of a multi-step thing, break down into 3-var units and stack) (Methodically attack _each part_ of the chain)
Analyzing and classifying every question stem ultimately determines the nature of the correct answer choice.
Analyzing and classifying every question stem ultimately determines the nature of the correct answer choice.
(Justify) If you've applied the Mechanistic Approach and still have to decide between contenders...
Apply the Justify Formula (Premises + Answer Choice == Conclusion)
(Formal) Additive Inf. Principle #6
Arrows and double-not (<-|->) arrows Because arrows and double-not arrows are so powerful, they almost always elicit additive inferences
(Common Flaws) False Dilemma
Assumes that only TWO (2) courses of action are available while there may be others
(Common Flaws) Time Shift Errors
Assuming that conditions will remain constant over time, and that what was the case in the past will be the case in the present or future
Assumption vs MBT Answers
Assumption answers contain statements that were USED TO MAKE the conclusion, whereas MBT answers contain statements that FOLLOW FROM the arg made in stimulus
To weaken a conditional conclusion...
Attack the Necessary condition by showing that the Necessary condition does NOT need to occur for the Sufficient condition to occur (Sufficient --> Necessary) (i.e. show that you can have the Sufficient without necessarily having the Necessary)
The key to weakening an LSAT argument is to...
Attack the conclusion! (Note that this doesn't always mean 'destroy', though that would be ideal)
(Common Flaws) Ad Hominem
Attacks the person (or source) instead of the argument they advance
(Common Flaws) Circular Reasoning
Author assumes as true what is supposed to be proved
(Common Flaws) False Analogy
Author uses an analogy that is TOO DISSIMILAR to the original situation to be applicable
(MoR) You must think about the STRUCTURE of the argument...
BEFORE examining the answer choices (i.e. not just the concrete facts of the argument)
In a Flaw question, you can identify the error of reasoning in the stimulus...
BEFORE proceeding to the answer choices
(MoR) IF you DO see the main conclusion at the end of a Method--AP problem...
Be prepared to answer a question about a part of the argument _other than_ the conclusion
(Assumption) If you encounter linked statements and an Assumption question...
Be prepared to supply the missing link or the contrapositive of that link
Premise Indicators
Because Since For For example For the reason that In that Given that As indicated by Due to Owing to This can be seen from We know this by (among others)
Counter-Premise Indicators
But Yet However On the other hand Admittedly In contrast Although Even though Still Whereas In spite of Despite After all
The phrase "must be false" is equivalent to...
CANNOT be true
(Parallel) When abstracting the flaw...
Describe the FLAW ITSELF, not how you'd fix it (b/c the right answer will parallel the flaw, NOT fill the flaw)
LR Primary Objective #1
Determine whether the stimulus contains an argument or if it is only a set of factual statements..
Remember to...
Diagram when it will help you better understand what the author is saying (i.e. use it sparingly! If you can see the conditional relationships clearly, then DON'T just diagram for the sake of diagramming) ("Typically, more advanced LSAT takers see the amount of diagramming they do _decrease_ over time")
"LSAT World" Scope Issues
Do NOT assume that everything you know "outside" of the stimulus regarding the topic is true and applies to the stimulus!
(Flaw) If you get to the end of the stimulus and are still super confused...
Do NOT proceed to the answer choices thinking that the answers will clarify or reveal the flaw! Keep hacking through it. (Many wrong answers are designed to subtly draw you toward side-issues)
The overarching LR argument question:
Do the given facts support the conclusion?
In Must Be True and Most Strongly Supported questions...
Don't bring in outside info and perceptions! The correct answer is provable/supported from the stimulus _alone_. (But you _can_ make 'commonsense' assumptions)
(PR) A principle ONLY tells you about the members of the group!
Don't widen the scope, even if it seems obvious!
(Formal) REMEMBER that "if and only if" is diagrammed by a...
Double-headed arrow (<-->)
(Formal) Remember that a SOME arrow is a ...
Double-headed arrow (<-s->)
(Common Flaws) Errors of Composition and Division
Errors that involve judgments made about groups and parts of a group 1.) An Error of Composition occurs when the author attributes a characteristic of PART of the group to the WHOLE group (or to each member) (i.e. PART to WHOLE) 2.) An Error of Division occurs when the author attributes a characteristic of the WHOLE (or each member of the whole) to a PART of the group (i.e. WHOLE to PART)
The "Uniqueness Rule of Answer Choices"
Every correct answer has a unique logical quality that meets the criteria in the question stem. Conversely, every incorrect answer has the OPPOSITE logical quality.
'Additional' Premise Indicators
Furthermore Moreover Besides In addition What's more After all
Additional Premise Indicators
Furthermore Moreover Besides In addition What's more After all
Link up the contrapositive in statements like (G --> H/) to form...
G <-|-> H
If you can't identify the conclusion (Main PT)
I think the Conclusion Identification Method(tm) is garbage. Use this instead: The Why-Because test: 1.) Why [conclusion]? 2.) Because [premise] If the arrangement make sense, you done good, else try flip
Parallel Reasoning questions ask you to...
Identify the answer choice that contains reasoning most similar in structure to the reasoning in the stimulus
Point of Agreement questions require you to...
Identify the issue or statement with which the two speakers would BOTH AGREE.
Assumption questions ask you to...
Identify what HAS to be true in order for the argument to work (i.e. an 'unstated premise')
Necessary Condition (Def)
If a necessary condition occurs, then it's possible but NOT CERTAIN that the sufficient condition will occur.
Sufficient Condition (Def)
If a sufficient condition occurs, you _automatically_ know that the necessary condition also occurs.
LR Primary Objective #9
If all 5 answer choices appear to be Losers, return to the stimulus and re-evaluate the argument.
(Main PT) Even if an answer choice must be true according to the stimulus...
If it fails to capture the main point, it cannot be correct
LR Primary Objective #3
If the stimulus contains an argument, determine whether the argument is strong or weak. (Do the premises strongly suggest that the conclusion would be true?)
Remember...
LEAST works like EXCEPT in question stems!
(Justify) The Mechanistic Approach (CONDENSED)
LINK NEW elements in the premises and conclusion AND IGNORE elements COMMON to both (Really powerful technique! Practice it on real questions until fluid!)
Most Strongly Supported
Like a Must Be True, but with a tad less certainty (correct answer might not be provable beyond a shadow of a doubt, though certainly could be)
When faced with an Assumption question, remember to...
Look for connections between rogue elements in the argument, and then seek that connection in the answer choices
(Formal) In making additive inferences, simply remember to...
Look for the weakest link in the chain and consider the relevant negativity
In LR, the word "inferred" means...
MUST be true
"Inference" in a LR question means...
MUST be true (What _follows_ from the argument)
When attempting to Resolve the paradox in the stimulus, you MUST address the facts of the situation...
Many incorrect answers will try to lure you with reasonable solutions that do _not quite meet_ the stated facts
If you get kinda stuck and can't sort our your conclusion from your premise[s]...
Mentally arrange them and see who depends upon who (i.e. the conclusion will depend upon the premise[s]) (In most cases, when you have the conclusion and premise backward, the arrangement will be confusing. The correct arrangement always sounds more logical)
The correct answer to an Assumption question involving Causality will normally fit into one of these categories...
Overarching theme: Assumptions that REINFORCE the causal relationship (cover all the bases). A.) Eliminates an alternate cause for the stated effect B.) Shows that when the cause occurs, the effect occurs C.) Shows that when the cause does not occur, the effect does not occur D.) Eliminates the possibility that the stated relationship is reversed E.) Shows that the data used to make the causal statement are accurate, or eliminates problems with the data
(MoR) Watch out for [incorrect] answers that are...
PARTIALLY true
A Justify question requires you to...
PROVE the conclusion (you are NOT asked to 'kind of' or 'somewhat' prove the conclusion)
(Weaken) A HUGE KEY to weakening arguments is to PERSONALIZE the argument
See it from your own perspective! i.e. how would you respond if you were talking directly to the author? (As opposed to trying to understand the issues abstractly)
(Eval.) In Evaluate the Argument questions, you must...
Select the answer choice that decides whether the argument is good or bad
LR Primary Objective #8
Separate the answer choices into Contenders and Losers. After completing this process, review the contenders and decide which answer is the correct one.
In a 'Weaken' question, the correct answer will most likely undermine the CONCLUSION by...
Showing that the conclusion fails to account for some element or possibility
LR Formal Logic: Quantity Terms (7Sage)
Some = 1-100 (can mean the same thing as most or all) Most = 51-100 (over half, can also mean all) All = 100 (referring to only that point) None = 0 Some, but not all = 1-99 (can mean some or most, but not all) Many = can include all!
(Formal) Additive Inf. Principle #9
Some and Most Combinations In general, 2 consecutive 'somes', 2 consecutive 'mosts', or a 'some' and 'most' in succession will not yield any inferences (remember the ladder and how it applies to stacking)
(Common Flaws) General Lack of Relevant Evidence for Conclusion
Some authors misuse info to such a degree that the fail to provide ANY info to support their conclusion, or they provide IRRELEVANT info
Do not be intimidated by the size of an LSAT question...
Some of the longest ones have been fairly easy!
(Formal) Additive Inf. Principle #1
Start by looking at the ends of the chain
Look carefully at what the author says about the topic at hand...
Statements presented as facts on the LSAT can and do vary from what occurs in the real world
Justify the Conclusion questions ask you to...
Strengthen the argument _so powerfully_ that the conclusion is 100% PROVEN
Strengthen questions ask you to...
Support the argument in any way possible, even if weakly
(Common Flaws) Exceptional Case/Overgeneralization (Sample Size Too Small)
Takes a small number of instances and treats those instances as if they support a broad, sweeping conclusion
(Parallel) Do not eliminate answers just because the wording of the conclusion is not identical [in certainty level or intent]...
Test makers like to use synonyms
(Causality) When an LSAT speaker concludes that one occurrence definitively caused another...
That speaker also assumes that the stated cause is the ONLY possible cause of the effect _and_ that consequently, the stated cause will ALWAYS produce the effect
(Parallel) When an answer has the same subject as the stimulus...
That's always a red flag!
(Formal) Additive Inf. Principle #5
The Most Train (Works much like the Some Train but b/c MOST is 1 step higher than SOME on the Logic Ladder, the Most Train produces _stronger_ inferences)
(Formal) Additive Inf. Principle #4
The SOME train To make an inference with a variable involved in a SOME relationship, either an ALL, NONE or <--> 'leading away' from the SOME relationship is required
(Assumption) If the evidence is NOT subjective...
The assumption will be!
In an LR conditional statement...
The conditional relationship stated by the author does NOT have to reflect reality.
In a Strengthen question...
The correct answer choice simply helps the argument in some way. The correct answer does not necessarily justify the argument, nor is the correct answer choice necessarily an assumption of the argument.
Remember...
The topic of a stimulus (however esoteric) does NOT affect the underlying logical relationship of the argument parts.
(Assumption) 'Supporter' Assumption
The traditional 'linking' assumption, i.e. what the author assumed to draw said conclusion from the premises (Closes the gap)
The double-NOT arrow only prohibits one scenario...
The two terms cannot occur together! (Remember, it does NOT exclude 'neither' from occurring)
(Formal) The "Logic Ladder"
The upper rungs IMPLY the lower rungs BUT the lower rungs DO NOT imply the upper rungs (they _might_ be true) All | | v Most | | v Some
(Formal) Additive Inf. Principle #2
The vast majority of additive inferences require either an ALL or NONE statement somewhere in the chain
If "except" is added to the question stem...
The logic is turned around/upside down. BUT it doesn't necessarily call for the polar opposite!!
MoR question stems refer to...
The method, technique, strategy, or process used by the author while making the argument
(MoR) Don't be intimidated by a bunch of showy language ('spinning swords')...
The right answer will at least sound like it's in the right ballpark.
Assumption answer choices CANNOT contain extraneous info because...
The statement must be something the author believed when forming the argument
(MoR) In Method--AP questions...
The stem cites a specific portion of the stimulus and then asks you to identify the role the _cited portion_ plays in the structure of the arg
Remember that...
The sufficient condition does not _make_ the necessary condition occur (it does NOT actively CAUSE the necessary condition to happen) Instead, it is an indicator that the necessary condition will occur, is occurring, or has already occurred
Must Be True (Mantra)
What can we PROVE?
(Assumption) In a typical argument...
There are an infinite number of assumptions possible, with most of those coming on the Defender side
"Only" and "Only if"
These, despite presence of "if", introduce NECESSARY conds (remember Sufficient --> Necessary)
When you see really esoteric terms...
They'll be defined somewhere, even if a few lines later
When creating an argument, an author will sometimes bring up a "counter-premise"...
This is a premise that actually contains an idea that is counter to the arg, and would generally weaken it.
Conclusion Indicators
Thus Therefore Hence Consequently As a result So Accordingly Clearly Must be that Shows that Conclude that Follows that For this reason (among others)
When "least" appears in a question stem...
Treat it EXACTLY THE SAME as "except"
(Causality) Diagramming a Causal Statement
Use arrows (like conditionals) BUT be careful! These are very different relationships than conditionals
(Formal) Additive Inf. Principle #7
Use inherent inferences
(Point at Issue) Don't overuse the Agree/Disagree Test!
Use it once you've narrowed things down to just a couple answer choices. Saves time.
(Common Flaws) Uncertain Use of a Term or Concept
Using a term in different ways/meanings is inherently confusing and undermines the integrity of the argument
(Formal) Additive Inf. Principle #8
Watch for the relevant negativity The presence of relevant negativity is defined as: 1.) Either the first or last term in sequence is negated 2.) There is a double-not arrow in the chain (_irrelevant_ negativity if negative is on the middle station)
(Formal) When we travel over MOST on the Most Train...
We get a ONE WAY pass
When a Parallel Reasoning stimulus contains flawed reasoning...
We ident it as a Parallel Flaw question (often it'll indicate in the stem that reasoning is flawed)
(Formal) Additive Inf. Principle #3
When making inferences, do NOT start with a variable involved in a double-not arrow (<-|->) relationship and then try to 'go across' the double-not arrow
(Justify) The Mechanistic Approach
When you [inevitably] get stuck on certain Justify questions, follow these rules: 1.) Any "new" element in the conclusion will appear in the correct answer 2.) Elements that are common to the conclusion and at least one premise, or to two premises, normally do NOT appear in the correct answer 3.) Elements in the premises but NOT the conclusion usually appear in the correct answer Practice this until fluid!
The secondary conclusion will support the main conclusion.
Why [MAIN CONCLUSION]? Because [SECONDARY CONCLUSION]
If a conclusion is present...
You MUST identify the conclusion prior to proceeding on to the question stem!
(Strengthen) Whenever you see a gap in the argument...
You can STRENGTHEN the argument by ELIMINATING that gap
(Parallel) If all else fails...
You can always fall back on describing the stimulus in abstract terms (less precise, but at least you get one last shot)
(Parallel) since each element must be matched...
You can analyze and attack the answer choices by testing whether the answer choice under consideration matches certain elements in the stimulus. If not, the answer is incorrect.
(Assumption) To negate a conditional statement...
You must show that the Necessary condition is NOT in fact NECESSARY (but could still be the case) (remember, Sufficient --> Necessary)
For "fill in the blank" Main PT questions...
_every time_, fill it in with an answer that best represents the main pt of the argument
(Flaw) Do not confuse a False dilemma with a situation where the author...
_legitimately_ establishes that only 2 possibilities exist (e.g. dead or alive)
The correct answer to a Main PT question will be...
a rephrasing of the main conclusion of the argument.
(MoR) Structurally, Method of Reasoning (MoR) questions are simply...
abstract MBTs
(PR) Principle (PR) questions are...
an "overlay" atop other question types (with "principle", "proposition", "precept", etc.)
(Parallel) Different methods can be used to eliminate different answers...
and the process should be FLUID and based on signals you derive from the stimulus
(Formal) The critical difference between the Some Train and Most Train is that...
because MOST has direction, you can ONLY _follow the MOST arrow_ to make a MOST inference
(Assumption) A Logical Opposite will always...
completely divide the subject under consideration into TWO (2) PARTS
Main PT question stems are remarkably consistent, asking you to ident the...
conclusion or point of the argument
Every main PT question stimulus...
contains an argument
(Formal) You can eke out a 'some' relationship from MOST and ALL arrows and use this to...
derive additive inferences (ride the Some Train)
(Main PT) Be careful with answers that are true according to the author...
do they also address the main point?
Temporally speaking...
either condition (Sufficient or Necessary) can occur first or at the same time
The stimulus for a Method of Reasoning question WILL contain an argument, and the argument can contain...
either valid or invalid reasoning (Make an assessment of the validity of the arg)
The stimuli for Weaken questions contain...
errors of ASSUMPTION (b/c the easiest argument to weaken is one that already has a flaw)
Most Must Be True and Most Strongly Supported stimuli are...
fact sets!
(Flaw) Beware of answers that describe a portion of the stimulus but...
fail to identify the error in the reasoning
Every stimulus in LSAT history prefaced by 'Advertisement:' has contained...
faulty or deceptive logic!
The LSAT always makes you pay if you...
gloss over ANY SECTION of the stimulus!
LSAC has shown that the average person understands that...
habitat size DOES affect population (A pretty extreme 'commonsense' assumption, I think)
(Parallel) The presence of a negative (-) term in the stimulus...
is NOT grounds for dismissing the answer when the stimulus has positive language (and vice-versa)
(Resolve) You must look VERY CLOSELY at the circumstances in the stimulus and...
make sure that the answer you select matches those circumstances
(Formal) In Formal Logic diagrams, each variable should appear...
only 1 time!
(Main PT) Any answer that ______ the MAIN conclusion of the argument will be the correct answer
paraphrases
To 'personalize' an argument, you can...
place yourself inside the situation and think how you'd react
The makers of the LSAT believe that surveys, _when conducted properly_...
produce reliable results
(Assumption) When the stimulus doesn't present any conditional chains, you should look for an answer that...
protects the necessary condition
In a Resolve question, you are NOT seeking to disprove one side of the situation. Instead, you are tasked with...
providing a POSSIBLE CAUSE of the paradoxical situation
(Flaw) If you ident a stimulus with causal reasoning and are asked a Flaw question...
quickly scan the answers for one that contains "cause", "effect", or both.
In Flaw in the Reasoning questions, the test makers...
tend to use certain types of answers again and again
(Formal) Double Arrow statements allow for only 2 possible outcomes...
the 2 vars occur TOGETHER, or NEITHER of the 2 vars occurs
(Formal) When writing out inferences, do NOT include...
the _inherent_ inferences. Write down the additive ones.
in a Strengthen Except question...
the four incorrect answers STRENGTHEN the argument and the _correct_ answer has either NO EFFECT or WEAKENS the arg
(Point at Issue) You can only use...
the info in the STIMULUS to evaluate the answer choices (i.e. much like MBT, no new stuff)
(Assumption) The Logical Opposite is the statement requiring...
the least amount of 'work' to negate the original statement (i.e. often just stick a NOT when not present, or remove one if so)
Look for modifiers like "only" and "all" (among others) because...
they may signal Exaggerated Answers (incorrect answers)
(Flaw) If you identify a stimulus with conditional reasoning and are asked a Flaw question...
you can quickly scan the answers for one answer that contains Sufficient, Necessary, or both.
(Eval) You are NOT being asked to prove with finality whether the argument is good or bad. Rather...
you must simply 'ask the question' that will best _analyze the validity_ of the argument
(Flaw) If you learn the mistakes that are often made by authors, then...
you will be able to quickly identify the error in the argument