Argument Extraction,Explanation,andEvaluation (EEE)

Most arguments that appearin philosophical writing, and in newspaper editorials, legal briefs, and ordinaryspeech, donot comewith numberedpremises and conclusions. But to evaluate such arguments, it is often helpful to first extract aformal argument, with numbered premises and conclusion, from thetext. It is often desirablealso to explain the resultingformal argument beforeevaluating it forvalidityand soundness.

1. Extracting Arguments withNumberedPremiseandConclusionfromTexts
To extractanargumentwith numbered premisesand conclusion from atext, do the following.

a. Locatethemainconclusion ofthe argumentand formulateit in clear, literal terminology.
b. Locatethecentral premises from which themainconclusion is derived, andstate them in clear, literal terminology.
c. Ifnecessary,add suppressed premises and subconclusions so as to makethe argument valid.
d. Eliminate all idlepremises.
e. Writethe entire argument in numbered premise-conclusion form.

Often manydifferent reasonable arguments can be extracted fromasingletext. Ourgoal when extractingarguments will beto find thestrongest,and most interesting, arguments that can reasonablybe attributed to the authoron thebasisofthetext.

1.1 Conclusion IndicatorWords
Somewords indicatea conclusion in thetext, including:

so, therefore, thus, hence, it follows that, consequently

Thesentences that appear afterthesewordsareconclusions (orsubconclusions)of arguments.

A. Fetuses areinnocent human beings, so abortion is wrong.
B. Words in English aremeaningful. Itfollows that thosewords havemeanings. C. No telescope, microscope, X-raymachine, PET scan, orMRIhaseverrevealed
thepresenceofameaninganywhere. Hence, there areno such thingsas meanings.

1.2 Premise IndicatorWords
Authors often indicatethepremises oftheirarguments with premiseindicator words. These include

because,for, since,given that

Thesewordsappearbeforethesentencethat is thepremise. Considerthefollowingverysimple text.

D. Abortion is wrongbecausefetuses areinnocent human beings. (Comparewith argument (A)above.)

Theword ‘because’ appears beforethesentence‘fetuses areinnocent human beings’. So this lattersentenceshould probablyappear as apremiseinyourextracted argument. Other examples:

E. Sincefetusesareinnocent human beings, takingtheirlives is morallywrong.
F. Meaningsareclearlymental in nature, for humanminds grasp themeanings of words.

1.3 Adding premises to obtain a valid argument
The extracted argument should bevalid: that is, each simpleargument in it should bevalid. An argument that is invalid can often betransformed into avalid argument thatthe authorwould accept byaddinga conditional premise. Forinstance, considerthesimpletext (G)below.

G. HillaryClinton is theU.S. SecretaryofState. So, even an idiot can seethat sheis a federal employee.

When extractinganumbered-premise-and-conclusion argument from (G),weleaveout the unnecessaryrhetoric. Asa first step weget (H)below.

H. 1. HillaryClinton is theU.S. SecretaryofState.
2. Therefore, HillaryClinton is a federal employee

This argument is formallyinvalid. But it can betransformed into avalid argument byadding a conditional premiseto get argument (I).

I. 1. HillaryClinton is theU.S. SecretaryofState.
2. IfHillaryClinton is theU.S. SecretaryofState, then HillaryClinton is a
federal employee.
3. Therefore, HillaryClinton is a federal employee.

Further, the added premiseis almost certainlyonethat the authorwould accept.

1.4 Eliminating Idlepremises
The argumentsyou extract from texts should haveno idlepremises. Informallyspeaking,an idle premiseis onethat playsno rolein supportingtheconclusion ofthe argument in which it
appears. A bit moreprecisely:

Def. S is an idlepremisein A iff: A is avalid argument,S is apremisein A, and the argument obtained bydeletingS from A is valid.

For example, sentence (3)in argument (J)is an idlepremisein (J).

J. 1. HillaryClinton is theU.S. SecretaryofState.
2. IfHillaryClinton is theU.S. SecretaryofState, then HillaryClinton is a
federal employee.
3. HillaryClinton receivesa check from theU.S. TreasuryDepartment every
week.
4. Therefore, HillaryClinton is a federal employee.

An even moreobvious caseof an idlepremiseis line (3)in (K).

K. 1. HillaryClinton is theU.S. SecretaryofState.
2. IfHillaryClinton is theU.S. SecretaryofState, then HillaryClinton is a
federal employee.
3. Snow is red.
4. Therefore, HillaryClinton is a federal employee.

Noticethat argument(K)is unsound becauseline(3)is false. Idlepremises arebad becausethey do nothingto support theargument’s conclusion, yet theyincreasethe riskthat the argument is unsound.

2. Explaining anExtractedArgument
To explainan argument thatyou have extracted, do the following.

a. Define all thetechnical terms that appearin theargument.
b. Give reasonsforeach ofthepremises oftheargument linebyline. (Do not give reasons forthemain conclusion and subconclusions.) In some cases, theauthor’s text provides reasons to believethepremise. In other cases,you must provide reasons whichwould lead a reasonableperson to accept thepremise. These should be consistent with theotherpremises oftheargument, and preferably reasons thatyou think the authorwould accept.

An explanation foralineis often, in effect, an argument forit. Sometimes an authorwill give reasons forapremise (oran argument forit)which, iftheywereincluded inthe formal numbered argument, will makethe formal argument impracticallylong. Inthis case,you can present whatyou taketobethemainreasons forthe conclusion in thenumbered premises and
put thesubsidiaryreasons in the explanation. In cases whereyourextracted argument contains a premise forwhich the authorprovides no reason,you should considerthe reasons that you can think of for acceptingit and present thosein the explanation.

Do not explainthe conclusions andsubconclusions ofarguments.Ifasimple argument is valid, and thepremises aretrue, then the conclusion must also betrue. So thereis no reason to explain the conclusion. Moreover, thetechnical terms that appearin conclusions ofsimple arguments in valid arguments usuallyappearin thepremises ofthosearguments (with the

exception ofsomeoddball arguments). So,you should have alreadydefined thoseterms when you explained thepremises.

3. Evaluating an ExtractedArgument
To evaluatean argumentthatyou have extractedand explained, do the following.

a. Statewhethereach simpleargument is valid orinvalid. Ifthe argument is complex, then for each simple argument in it, statewhetherthat simple argument is valid. (Ifyou extracted the argument correctly,each simpleargument will be valid.)
b. Foreach simple argument that is valid, statethenameofthelogical form that it exemplifies (MP, MT, etc.).
c. Ifthe argument has oneormorepremises thatarefalseor controversial, point out theweakest such premiseand present someobjection to it. Besureto specify which premiseyou areobjectingto.
d. Statewhethertheargument is sound. You mayconcludethat the argument is sound, despitetheobjectionyou presented to oneofits premises. In that case, you might want to present a responseto theobjection, ifyou havespace.

Iwill usuallyaskyou tostatea reasonableobjection to the argument, evenifyou think that it is sound. Sometimes,Iwill askyou to describehowthe authormight respondto the criticismyou present in (c).

4. SampleTexts forEEE

1. Philosophers who accept the referential theoryofmeaningsaythat themeaningofaword is thethingit refers to. Butthat’s just silly. Look, theword ‘the’has ameaning–it’s not just a meaningless sound. Butjust trypointingat thethingthat it refers to!Obviously, ‘the’doesn’t referto anything. Thosereferentialists need to learn alittle common sense.

2. Georgeis suchaloser. Hejust contradicted himself!Hesaid“Batmandoes not exist”. But when Georgeused thename ‘Batman’hereferredto Batman. (Who elsewould hebe referring to? BillClinton,maybe?)Butyou can’t refertoBatman ifBatman doesn’t exist. So Batman exists.Which just goes to show: Georgeis atotaldork.

3. Anyonewho puts asidehis prejudices and just thinks foraminutewill realizethat there are nonexistent objects. Look, obviouslyHomerSimpson, SantaClaus, and HuckleberryFinn are things don’t exist. So, therearethings that don’t exist, and so there arenonexistent objects. Yes, Virginia, thereis aSanta Claus, onlyhedoesn’t exist.

4. Ifmeanings areideas, then all we evertalk about areourown ideas. How stupid is that?

5. An ExampleofHow to Extract, Explain, and EvaluateanArgument
Let’s work throughan example. Consider argument (1) above.

Philosophers who accept the referential theoryofmeaningsaythat themeaningofaword is thethingit refers to. But that’s just silly. Look,theword ‘the’has ameaning–it’s not just ameaningless sound. But just trypointing at thethingthat it refers to! Obviously,
‘the’doesn’t referto anything. Thosereferentialists need to learnalittle common sense.

When the authorsays ‘But that’s just silly’, he/sheseems to becriticizingphilosophers (certain people)who accept thereferential theoryofmeaning. But wearenot interested in arguments that criticizepeople. We areinstead interested inarguments that criticizetheories, orclaims. So, the argument wewillextract will criticizethe referential theory. Moreprecisely, the
argument that weextract will haveamain conclusion that says that thereferential theoryis false oruntrue. And so areasonable conclusion forour extracted argument would be:

The referential theoryofmeaningis false.

Themain evidencegivenin favorofthis conclusion seems to be: Theword ‘the’doesn’treferto anything.
So, wehave an initial structureto start.

Theword ‘the’doesn’treferto anything.
Therefore, thereferentialtheoryofmeaningis false.

This bit of argument is notyet formallyvalid. But we could makeit valid byadding a conditional (if-then)premiseto obtain avalid argument, as follows.

Theword ‘the’doesn’treferto anything.
Iftheword‘the’doesn’t’referto anything, then the referential theoryofmeaning is false.
Therefore, thereferentialtheoryofmeaningis false.

This argument is an exampleofModus Ponens. We could instead extract the following argument, which has theform ofModus Tollens, from thetext.

Ifthe referential theoryofmeaningis true, then ‘the’ refers to something. It is not thecasethat theword ‘the’refers to something.
Therefore, thereferentialtheoryofmeaningis nottrue.

Noticethat whenI rephrasethe argument in this way, thesecond lineis exactlythenegation of the ‘then’part (the consequent)ofthe first line. Isomewhat preferthis lastextracted argument, because(i)it has the form ofMT and manyarguments against theories takethat form and (ii)its first premisemakesclearthat we aredealingwitha consequenceofthe referential theory.

Let’s takethis to beourextracted argument. Idon’t likethis extraction much, becauseit leaves out so much ofwhat theauthorsays, particularlythe reasons that theauthorpresents for acceptingthepremises ofthe aboveargument. But we can includethesereasons in our explanation ofthe argument. Hereis howwemight do it on an exam.

Exam Question: Extract an argument with numbered premisesand conclusions from the followingpassage.

Philosophers who accept the referential theoryofmeaningsaythat themeaningofaword is thethingit refers to. But that’s just silly. Look,theword ‘the’has ameaning–it’s not just ameaningless sound. But just trypointing at thethingthat it refers to! Obviously,
‘the’doesn’t referto anything. Thosereferentialists need to learnalittle common sense.

Answer:

Extracted Argument:
1. Ifthe referential theoryofmeaningis true, then ‘the’ refers to something.
2. It is not thecasethat theword ‘the’refers to something.
3. Therefore, thereferentialtheoryofmeaningis nottrue. [from 1, 2]

Explanation:
(1) Assumethat the referential theoryofmeaningis true. Thereferential theorysays that themeaningof anymeaningful expression is thethingto which it refers. The word ‘the’is not ameaningless sound. So, ‘the’is ameaningful expression. So,
it has ameaning. So (assumingthat the referentialtheoryofmeaningis true)its meaningis thethingto which is refers. So, ‘the’refers to its meaning, andso
‘the’ refers to something. Therefore, ifthereferential theoryofmeaningistrue, then ‘the’refers to something. Therefore, line (1)is true.
(2) If‘the’referred to something, then someonewould be ableto point at thethingto which it refers. But ‘the’clearlydoes not refertoanyparticular rock, ortable, or person. In fact, it’s hardto think of anythingat which we could point thatwould bea reasonablecandidate forthethingthat ‘the’refers to. So, no onecanpoint at an object that ‘the’ refersto. Therefore, we can inferthat (2)is true: It is not the casethat theword ‘the’refers to something.

Evaluation:
Validity: 1, 2 =>: Valid, byMT
Soundness: Thesecondpremiseis themost vulnerableto criticism. It could be claimed that theword ‘the’does referto somethingbut thethingto which it refers is something ratherabstract,and so not somethingthat onecanpoint at.
However, in myopinion, thereareno suchabstract objects that one cannot point at. So
I concludethat theargument is unsound.

End of example. NoticethatIexplained each premiseofthe extractedargument. My explanation of each lineconsisted of an argumentfor that premise. In thiscase, thearguments that appearin theexplanation arejust slightlymodified versions ofwhat the authorwrote.

Notice also thatIdid notexplain the conclusion. (Also notethat the aboveargument and explanation would beveryhard to follow ifIdid not put quotation marks around ‘the’whenI was writing about that word.)Noticethat whenI evaluated forvalidity,Iexplicitlysaid that the argument was valid, andIstated the argument-form that the extracted argument instantiates.

Ithen presented areasonableobjection to the argument. Noticethat whenIdid soIexplicitly statedwhich premiseIwas criticizing. However,Ithen went on to rejectthat criticism, and said why,andIconcluded that the argument is sound. (Bytheway,Ido not endorsethe criticism orthe responseto the criticism.)

An AlternativeExtraction

Therearemanyreasonable arguments with numbered premisesand conclusion that one can extract from the abovetext. Hereis anotherone,whichIpreferto theoneabove, becauseit gives moreinformation in the explicit premises.

1. Ifthe referential theoryofmeaningis true, then ‘the’ refers to something.
2. If‘the’refers to something, then someone couldpoint at thethingto which ‘the’
refers.
3. Therefore, ifthereferential theoryofmeaningis true, then someone could point at thethingto which ‘the’refers. [from 1, 2]
4. It is not thecasethat someone could point at thethingto which ‘the’ refers.
5. Therefore, thereferentialtheoryofmeaningis nottrue. [from 3, 4]

Explanation:
(1) [Same as explanation as for (1)in the above example]
(2) Assumethat ‘the’refers to something. We can point at anythingthat anyword can referto. So, if ‘the’refers to something, thensomeone could point at the thingto which ‘the’ refers. Therefore (2)is true.
(3) Subconclusion.
(4) Consider all ofthethingswe can point at: rocks, table, stars, people,cats, and so on. Clearly‘the’does not refertoanyofthem. Therefore, ‘the’does not referto somethingthat wecan point at. Therefore, it is not the casethat someone could point at thethingto which ‘the’ refers.

Evaluation:
This is a complexargument, with two subarguments.
1, 2 =>3: Valid, Hypothetical Syllogism
3, 4 =>5: Valid, ModusTollens

Each subargument in this complexargument is valid. Therefore, this argument is sound iflines (1), (2)and(3) are all true.
A criticmight claim that line (2)is false. Sucha critic could claim that somewords referto things that we can’t point at. Perhaps theword ‘justice’is suchaword: ‘we can’t reallypoint at justiceitself. Ifso, then perhaps ‘the’ also refers to somethingthat we can’t point at.

In myopinion, however, this criticism ofline(2)is faulty,forthere areno abstract things likejustice. Everythingthat is real is concrete and can bepointedat. Iconclude that the argument is sound.

End of example. NoticethatIam not endorsingthe criticism ofthe aboveargument, orthe responseto thecriticism. Iam presentingthecriticism and responseforillustrativepurposes only.
PLACE THIS ORDER OR A SIMILAR ORDER WITH US TODAY AND GET AN AMAZING DISCOUNT