week 12. language universals, and the beginnings of a model grs lx 700 language acquisition and...

Week 12.Week 12.Language Universals, and the Language Universals, and the

beginnings of a modelbeginnings of a model

GRS LX 700GRS LX 700Language Language

Acquisition andAcquisition andLinguistic TheoryLinguistic Theory

Typological universalsTypological universals

1960’s and 1970’s saw a lot of 1960’s and 1970’s saw a lot of activity aimed at identifying activity aimed at identifying language universalslanguage universals, properties of , properties of Language.Language.

Class of Class of possible languagespossible languages is is smaller than you might think.smaller than you might think.

If a language has one property (A), it If a language has one property (A), it will necessarily have another (B).will necessarily have another (B). +A+A+B+B, , –A–B–A–B, , –A–A+B+B but neverbut never +A+A–B–B..

(Typological) universals(Typological) universals

All languages have vowels.All languages have vowels.

If a language has VSO as its basic word If a language has VSO as its basic word order, then it has order, then it has preprepositions (vs. positions (vs. postpostpositions).positions).

VSO?VSO?

Adposition typeAdposition typeYes Yes NoNo

PrePrepositionspositions WelshWelsh EnglishEnglish

PostPostpositionspositions NoneNone JapanesJapanesee

MarkednessMarkedness

Having Having dualsduals implies having implies having pluralsplurals Having Having plurals plurals says nothing about having says nothing about having

dualsduals..

Having duals is Having duals is markedmarked—infrequent, more —infrequent, more complex. Having plurals is (relative to complex. Having plurals is (relative to having duals) having duals) unmarkedunmarked..

Generally markedness is in terms of Generally markedness is in terms of comparable dimensions, but you comparable dimensions, but you couldcould also also say that say that being VSO being VSO is marked relative to is marked relative to having prepositionshaving prepositions..

MarkednessMarkedness

““Markedness” actually has been used Markedness” actually has been used in a couple of different ways, although in a couple of different ways, although they share a common core.they share a common core.

Marked:Marked: More unlikely, in some sense. More unlikely, in some sense. Unmarked:Unmarked: More likely, in some More likely, in some

sense.sense.

You have to “mark” something marked; You have to “mark” something marked; unmarked is what you get if you don’t unmarked is what you get if you don’t say anything extra.say anything extra.

““Unlikeliness”Unlikeliness” Typological/crosslinguistic Typological/crosslinguistic infrequencyinfrequency..

VOS word order is marked.VOS word order is marked. More complexMore complex constructions. constructions.

[ts] is more marked than [t].[ts] is more marked than [t]. The The non-default settingnon-default setting of a parameter. of a parameter.

Non-null subjects?Non-null subjects? Language-specificLanguage-specific//idiosyncraticidiosyncratic

features.features. Vs. UG/universal features…?Vs. UG/universal features…?

Berlin & Kay 1969: Color Berlin & Kay 1969: Color termsterms

(On the boundaries of (On the boundaries of psychophysics, linguistics, psychophysics, linguistics, anthropology, and with issues about anthropology, and with issues about its its interpretationinterpretation, but still…), but still…)

Basic color terms Basic color terms across languages.across languages. It turns out that languages differ in It turns out that languages differ in

how many color terms count as how many color terms count as basic.basic. ( (blueishblueish, , salmon-coloredsalmon-colored, , crimsoncrimson, , blondblond, … are not basic)., … are not basic).


The segmentation of experience by speech The segmentation of experience by speech symbols is essentially arbitrarysymbols is essentially arbitrary. The different sets . The different sets of words for color in various languages are of words for color in various languages are perhaps the best ready evidence for such essential perhaps the best ready evidence for such essential arbitrariness. For example, in a high percentage of arbitrariness. For example, in a high percentage of African languages, there are African languages, there are only three “color only three “color words,” corresponding to our white, black, red, words,” corresponding to our white, black, red, which nevertheless divide up the entire spectrum.which nevertheless divide up the entire spectrum. In the Tarahumara language of Mexico, there are In the Tarahumara language of Mexico, there are five basic color words, and here “blue” and five basic color words, and here “blue” and “green” are subsumed under a single term“green” are subsumed under a single term.. Eugene Nida (1959)Eugene Nida (1959)


ArabicArabic (Lebanon) (Lebanon) BulgarianBulgarian (Bulgaria) (Bulgaria) CatalanCatalan (Spain) (Spain) CantoneseCantonese (China) (China) MandarinMandarin (China) (China) EnglishEnglish (US) (US) HebrewHebrew (Israel) (Israel) HungarianHungarian (Hungary) (Hungary) IbiboIbibo (Nigeria) (Nigeria) IndonesianIndonesian

(Indonesia)(Indonesia)

JapaneseJapanese (Japan) (Japan) KoreanKorean (Korea) (Korea) PomoPomo (California) (California) SpanishSpanish (Mexico) (Mexico) SwahiliSwahili (East Africa) (East Africa) TagalogTagalog (Philippines) (Philippines) ThaiThai (Thailand) (Thailand) TzeltalTzeltal (Southern (Southern

Mexico)Mexico) UrduUrdu (India) (India) VietnameseVietnamese (Vietnam) (Vietnam)

Eleven possible basic Eleven possible basic color termscolor terms

White, black, red, green, yellow, blue, White, black, red, green, yellow, blue, brown, purple, pink, orange, gray.brown, purple, pink, orange, gray.

All languages contain term for white and All languages contain term for white and black.black.

Has Has 33 terms, contains a term for terms, contains a term for redred.. Has Has 44 terms, contains terms, contains green green or or yellowyellow.. Has Has 55 terms, contains terms, contains both both green green andand yellowyellow.. Has Has 66 terms, contains terms, contains blueblue.. Has Has 77 terms, contains terms, contains brownbrown.. Has Has 8 or more 8 or more terms, chosen from {terms, chosen from {purplepurple, ,

pinkpink, , orangeorange, , graygray}}

Color hierarchyColor hierarchy White, blackWhite, black RedRed Green, yellowGreen, yellow BlueBlue BrownBrown Purple, pink, orange, grayPurple, pink, orange, gray

Even assuming these 11 basic color terms, there Even assuming these 11 basic color terms, there should be 2048 possible sets—but only 22 (1%) should be 2048 possible sets—but only 22 (1%) are attested.are attested.

Color termsColor terms

BWBW Jalé (New Guinea) ‘brilliant’ vs. ‘dull’Jalé (New Guinea) ‘brilliant’ vs. ‘dull’ BWBWRR Tiv (Nigeria), Australian aboriginals inTiv (Nigeria), Australian aboriginals in

Seven Rivers District, Queensland.Seven Rivers District, Queensland. BWBWRRGG Ibibo (Nigeria), Hanunóo (Philippines)Ibibo (Nigeria), Hanunóo (Philippines) BWBWRRYY Ibo (Nigeria), Fitzroy River people (Queensland)Ibo (Nigeria), Fitzroy River people (Queensland) BWBWRRYYGG Tzeltal (Mexico), Daza (eastern Nigeria)Tzeltal (Mexico), Daza (eastern Nigeria) BWBWRRYYGGUU Plains Tamil (South India), Nupe (Nigeria), Plains Tamil (South India), Nupe (Nigeria),

Mandarin?Mandarin? BWBWRRYYGGUUOO Nez Perce (Washington), Malayalam (southern Nez Perce (Washington), Malayalam (southern

India)India)

Color termsColor terms

Interesting questions abound, including Interesting questions abound, including whywhy this order, this order, whywhy these eleven—and these eleven—and there are potential reasons for it that there are potential reasons for it that can be drawn from the perception of can be drawn from the perception of color spaces which we will not attempt color spaces which we will not attempt here.here.

The point is: The point is: This is a fact about This is a fact about Language: Language: If you have a basic color term If you have a basic color term for for blueblue, you also have basic color terms , you also have basic color terms for for blackblack, , whitewhite, , redred, , greengreen, and , and yellowyellow..

Implicational hierarchyImplicational hierarchy

This is a This is a ranking of markednessranking of markedness or an or an implicational hierarchyimplicational hierarchy..

Having Having blueblue is more marked than having (any is more marked than having (any or all of) or all of) yellowyellow, , greengreen, , redred, , whitewhite, and , and blackblack..

Having Having greengreen is more marked than having is more marked than having redred……

Like a set of implicational universals…Like a set of implicational universals… Blue implies yellowBlue implies yellow Brown implies blueBrown implies blue Blue implies greenBlue implies green Pink implies brownPink implies brown Yellow or green imply red Yellow or green imply red Orange implies brownOrange implies brown Red implies blackRed implies black Gray implies brownGray implies brown Red implies whiteRed implies white Purple implies brownPurple implies brown

L2A?L2A?

Our overarching theme:Our overarching theme:How much is L2/IL like a L1?How much is L2/IL like a L1?

Do IL/L2 languages obey the Do IL/L2 languages obey the language universals that hold of language universals that hold of native languages?native languages?

This question is slightly less This question is slightly less theorytheory--laden than the questions we were laden than the questions we were asking about principles and asking about principles and parameters, although it’s similar…parameters, although it’s similar…

To my knowledge nobody has studied To my knowledge nobody has studied L2 acquisitions of color terms…L2 acquisitions of color terms…

Question formationQuestion formation

Declarative: Declarative: John will buy coffee.John will buy coffee.

WhWh-inversion: -inversion: What What willwill JohnJohn buy?buy? WhWh-fronting: -fronting: WhatWhat will John buy?will John buy? Yes/No-inversion: Yes/No-inversion: WillWill JohnJohn buy coffee?buy coffee?

Greenberg (1963):Greenberg (1963): WhWh-inversion implies -inversion implies WhWh-fronting.-fronting. Yes/No-inversion implies Yes/No-inversion implies WhWh-inversion.-inversion.

WhWh-inversion-inversionWhWh--frontingfronting

English, German:English, German: Both. Both. WhatWhat willwill JohnJohn buy? buy?

Japanese Korean:Japanese Korean: neither.neither. John will buy what?John will buy what?

Finnish:Finnish: WhWh-fronting only.-fronting only. WhatWhat John will buy? John will buy?

UnattestedUnattested:: WhWh-inversion -inversion only.only. *Will John buy what?*Will John buy what?

Y/N-inversionY/N-inversionWhWh--inversioninversion

English:English: BothBoth WillWill JohnJohn buy coffee? What buy coffee? What willwill JohnJohn buy? buy?

Japanese:Japanese: NeitherNeither John will buy coffee? John will buy what?John will buy coffee? John will buy what?

Lithuanian:Lithuanian: WhWh-inversion only.-inversion only. John will buy coffee? What John will buy coffee? What willwill JohnJohn buy? buy?

UnattestedUnattested:: Y/N-inversion only.Y/N-inversion only. WillWill JohnJohn buy coffee? What John will buy? buy coffee? What John will buy?

Eckman, Moravcsik, Eckman, Moravcsik, Wirth (1989)Wirth (1989)

L1: Korean (4), Japanese (6), Turkish L1: Korean (4), Japanese (6), Turkish (4)(4)

L2: EnglishL2: English

Note L1s chosen because they are Note L1s chosen because they are neither/neither type languages, to neither/neither type languages, to avoid questions of transfer.avoid questions of transfer.

Subjects tried to determine what was Subjects tried to determine what was going on in a scene by asking going on in a scene by asking questions.questions.


Example Y/N Qs:Example Y/N Qs: DidDid sheshe finished two bottle wine? finished two bottle wine? IsIs Lou and PattyLou and Patty known each other? known each other? Sue does drink orange juice?Sue does drink orange juice? Her parents are rich?Her parents are rich? IsIs this storythis story is chronological in a order? is chronological in a order? DoesDoes JoanJoan has a husband? has a husband? Yesterday Yesterday isis SueSue did drink two bottles did drink two bottles

of wine?of wine?


Example Example WhWh-Qs:-Qs: WhyWhy Sue didn’t look solution for her Sue didn’t look solution for her

problem?problem? WhereWhere Sue is living? Sue is living? WhyWhy diddid SueSue stops drinking? stops drinking? WhyWhy isis PattyPatty’s going robbing the bank?’s going robbing the bank? WhatWhat they are radicals? they are radicals? WhatWhat Sue and Patty connection? Sue and Patty connection? WhyWhy she was angry? she was angry?

Eckman Eckman et al. et al.

(1989)(1989)

whwh-inv-invwhwh--

fronting?fronting?

resultsresults

%% WhinvWhinv %% WhfrWhfr

SMSM KK 2525 NONO 100100 YESYES

UAUA TT 5454 NONO 100100 YESYES

TSTS JJ 7070 NONO 100100 YESYES

MKMK KK 8080 NONO 100100 YESYES

RORO JJ 8888 NONO 100100 YESYES

KOKO JJ 9595 YESYES 100100 YESYES

MMHH

JJ 9595 YESYES 100100 YESYES

NENE TT 9595 YESYES 100100 YESYES

SISI JJ 9595 YESYES 100100 YESYES

GG TT 100100 YESYES 100100 YESYES

MAMA TT 100100 YESYES 100100 YESYES

STST JJ 100100 YESYES 100100 YESYES

TMTM KK 100100 YESYES 100100 YESYES

YKYK JJ 100100 YESYES 100100 YESYES

Eckman Eckman et al. et al.

(1989)(1989)

YN-inv.YN-inv. whwh-inv.?-inv.?

resultsresults

%% YNinvYNinv %% WHinvWHinv

SMSM KK 88 NONO 2525 NONO

MKMK KK 3838 NONO 8080 NONO

YKYK JJ 5151 NONO 101000

YESYES

TSTS JJ 6767 NONO 7070 NONO

TMTM KK 8383 NONO 101000

YESYES

RORO JJ 8585 NONO 8888 NONO

BGBG TT 8686 NONO 101000

YESYES

MAMA TT 8888 NONO 101000

YESYES

UAUA TT 9191 YESYES 5454 NONO

KOKO JJ 9393 YESYES 9595 YESYES

MHMH JJ 9595 YESYES 9595 YESYES

NENE TT 100100 YESYES 9595 YESYES

SISI JJ 100100 YESYES 9595 YESYES

STST JJ 100100 YESYES 101000

YESYES


Yes/no Yes/no inversioninversion

Wh-inversionWh-inversion

Yes (VS)Yes (VS) No (SV)No (SV)

Yes (VS)Yes (VS) 55 44

No (SV)No (SV) 11 44

Eckman’s Markedness Eckman’s Markedness Differential HypothesisDifferential Hypothesis

Markedness. Markedness. A phenomenon or structure X in A phenomenon or structure X in some language is some language is relatively more markedrelatively more marked than than some other phenomenon or structure Y if cross-some other phenomenon or structure Y if cross-linguistically the presence of X in a language linguistically the presence of X in a language implies the presence of Y, but the presence of Y implies the presence of Y, but the presence of Y does not imply the presence of X.does not imply the presence of X.

DualsDuals imply plurals. imply plurals. WhWh-inversion-inversion implies implies whwh-fronting.-fronting. BlueBlue implies red. implies red.

Markedness Differential Markedness Differential HypothesisHypothesis

MDH: The areas of difficulty that a second MDH: The areas of difficulty that a second language learner will have can be predicted language learner will have can be predicted on the basis of a comparison of the NL and on the basis of a comparison of the NL and TL such that:TL such that: Those areas of the TL that are different from the Those areas of the TL that are different from the

NL and are relatively more marked than in the NL and are relatively more marked than in the NL will be difficult;NL will be difficult;

The degree of difficulty associated with those The degree of difficulty associated with those aspects of the TL that are different and more aspects of the TL that are different and more marked than in the NL corresponds to the marked than in the NL corresponds to the relative degree of markedness associated with relative degree of markedness associated with those aspects;those aspects;

Those areas of the TL that are different than the Those areas of the TL that are different than the NL but are not relatively more marked than in NL but are not relatively more marked than in the NL will not be difficult.the NL will not be difficult.

MDH example:MDH example:Word-final segmentsWord-final segments

Voiced obstruentsVoiced obstruents most markedmost marked SurgeSurge Voiceless obstruentsVoiceless obstruents CokeCoke Sonorant consonantsSonorant consonants MountainMountain VowelsVowels least markedleast marked CoffeeCoffee

All Ls allow vowels word-finally—some All Ls allow vowels word-finally—some onlyonly allow allow vowels. Some (e.g., vowels. Some (e.g., MandarinMandarin, , JapaneseJapanese) allow ) allow only vowels and sonorants. Some (e.g., only vowels and sonorants. Some (e.g., PolishPolish) ) allow vowels, sonorants, but only allow vowels, sonorants, but only voicelessvoiceless obstruents. obstruents. EnglishEnglish allows all four types.allows all four types.

Eckman (1981)Eckman (1981)

Spanish L1Spanish L1 Mandarin L1Mandarin L1

GlossGloss IL formIL form GlossGloss IL formIL form

BobBob [b p][b p] TagTag [tæg ][tæg ]

BobbyBobby [b bi][b bi] AndAnd [ænd ][ænd ]

RedRed [r[rt]t] WetWet [w t][w t]

WetWet [w t][w t] DeckDeck [d[dk]k]

SickSick [s[sIIk]k] LetterLetter [l[lt r]t r]

BleedingBleeding [blid[blidIIn]n]

c

c

e

e

e

e

e

MDH example:MDH example:Word-final segmentsWord-final segments


Idea: Idea: MandarinMandarin has neither voiceless nor voiced has neither voiceless nor voiced obstruents in the L1—using a voiceless obstruent in obstruents in the L1—using a voiceless obstruent in place of a TL voiced obstruent is still not L1 compliant place of a TL voiced obstruent is still not L1 compliant and is a big markedness jump. Adding a vowel is L1 and is a big markedness jump. Adding a vowel is L1 compliant. compliant. SpanishSpanish has voiceless obstruents, to using a has voiceless obstruents, to using a voiceless obstruent for a TL voiced obstruent is L1 voiceless obstruent for a TL voiced obstruent is L1 compliant.compliant.

MDH and ILMDH and IL

The MDH presupposes that the IL The MDH presupposes that the IL obeys the implicational universals too.obeys the implicational universals too.

Eckman et al. (1989) suggests that Eckman et al. (1989) suggests that this is at least reasonable.this is at least reasonable.

The MDH suggests that there is a The MDH suggests that there is a natural order of L2A along a natural order of L2A along a markedness scale (stepping to the markedness scale (stepping to the next level of markedness is easiest).next level of markedness is easiest).

Let’s consider what it means that an Let’s consider what it means that an IL obeys implicational universals…IL obeys implicational universals…

MDH and ILMDH and IL

IL obeys implicational universals.IL obeys implicational universals. That is, That is, we know that IL is a languagewe know that IL is a language.. So, we know that languages are such So, we know that languages are such

that having word-final voiceless that having word-final voiceless obstruents implies that you also have obstruents implies that you also have word-final sonorant consonants, among word-final sonorant consonants, among other things.other things.

What would happen if we taught What would happen if we taught Japanese L2 learners of English only—Japanese L2 learners of English only—and at the outset—voiced obstruents?and at the outset—voiced obstruents?

Generalizing with Generalizing with markedness scalesmarkedness scales


JapaneseJapanese learner of learner of EnglishEnglish will have an easier will have an easier time at each step learning voiceless obstruents time at each step learning voiceless obstruents and then voiced obstruents.and then voiced obstruents.

But—if taught voiced obstruents immediately, the But—if taught voiced obstruents immediately, the fact that the IL obeys implicational (markedness) fact that the IL obeys implicational (markedness) universals means that voiceless obstruents “come universals means that voiceless obstruents “come for free.”for free.”

NiftyNifty!!

Does it work? Does it help?Does it work? Does it help? Answers seem to be:Answers seem to be:

Yes, it seems to at least sort of work.Yes, it seems to at least sort of work. MaybeMaybe it helps. it helps.

Learning a marked structure is Learning a marked structure is harderharder.. So, if you learn a marked structure, So, if you learn a marked structure, you can automatically generalize to the you can automatically generalize to the less marked structuresless marked structures, but , but was it was it faster than learning the easier steps in faster than learning the easier steps in succession would have been?succession would have been?

Change from pre- to Change from pre- to post-testpost-test

Eckman, Bell, & Nelson Eckman, Bell, & Nelson (1988)(1988)

The Noun Phrase The Noun Phrase Accessibility HierarchyAccessibility Hierarchy

Keenan & Comrie (1977) observed a hierarchy Keenan & Comrie (1977) observed a hierarchy among the kinds of relative clauses that languages among the kinds of relative clauses that languages allow.allow.

The astronaut [(that) I met yesterday].The astronaut [(that) I met yesterday]. Head noun:Head noun: astronautastronaut Modifying clause:Modifying clause:

((that/who) I met — yesterday.that/who) I met — yesterday. Compare:Compare: I met I met the astronautthe astronaut yesterday. yesterday. This is an This is an object relativeobject relative because the place where because the place where

the head noun the head noun wouldwould be in the simple sentence be in the simple sentence version is the object.version is the object.


There are several kinds of relative clauses, There are several kinds of relative clauses, based on where the head noun “comes from” based on where the head noun “comes from” in the modifying clause:in the modifying clause:

The astronaut…The astronaut… [I met — yesterday][I met — yesterday] objectobject [who — met me yesterday][who — met me yesterday] subjectsubject [I gave a book to —][I gave a book to —] indirect objectindirect object [I was talking about —][I was talking about —] obj. of Pobj. of P [whose house I like —][whose house I like —] Genitive (possessor)Genitive (possessor) [I am braver than —][I am braver than —] obj. of obj. of

comparativecomparative


Turns out:Turns out: Languages differ in what Languages differ in what positions they allow relative clauses to positions they allow relative clauses to be formed on.be formed on.

EnglishEnglish allows allows all the positions all the positions mentioned to be used to make relative mentioned to be used to make relative clauses.clauses.

ArabicArabic allows relative clauses to be allows relative clauses to be formed only with formed only with subjectssubjects..

GreekGreek allows relative clauses to be allows relative clauses to be formed only with formed only with subjects subjects or or objectsobjects..

Resumptive pronounsResumptive pronouns

The guy The guy who they don’t know whether who they don’t know whether hehe wants to come.wants to come.

A student A student who I can’t make any sense out who I can’t make any sense out of the papers of the papers hehe writes. writes.

The actress The actress who Tom wondered whether who Tom wondered whether her her father was rich.father was rich.

In cases where relative clause formation is In cases where relative clause formation is notnot allowed, it can sometimes be salvaged allowed, it can sometimes be salvaged by means of a pronoun in the position that by means of a pronoun in the position that the head noun is to be associated with.the head noun is to be associated with.

NPAH and resumptive NPAH and resumptive pronounspronouns

Generally speaking, it turns out that in languages Generally speaking, it turns out that in languages which which do notdo not allow relative clauses to be formed allow relative clauses to be formed off a certain position, they will off a certain position, they will insteadinstead allow allow relative clauses with a resumptive pronoun in that relative clauses with a resumptive pronoun in that position.position.

Arabic:Arabic: allows only subject relative clauses. But allows only subject relative clauses. But for all other positions allows a resumptive pronoun for all other positions allows a resumptive pronoun construction, analogous to:construction, analogous to: The book The book that John bought that John bought itit.. The tree The tree that John is standing by that John is standing by itit.. The astronaut The astronaut that John gave that John gave him him a present.a present.

NPAHNPAH

The positions off which you can relativize The positions off which you can relativize appears to be an appears to be an implicational hierarchyimplicational hierarchy..

Lang.Lang. SUBSUB DODO IOIO OPOP GENGEN OCOMPOCOMP

ArabicArabic –– ++ ++ ++ ++ ++

GreekGreek –– –– +?+? +?+? ++ ++

JapanesJapanesee

–– –– –– –– +/ +/ ––

PersianPersian –– (+)(+) ++ ++ ++ ++

Noun Phrase Noun Phrase Accessibility HierarchyAccessibility Hierarchy

More generally, there seems to be a More generally, there seems to be a hierarchy of “difficulty” (or hierarchy of “difficulty” (or “(in)accessibility”) in the types of “(in)accessibility”) in the types of relative clauses.relative clauses.

A language which allows this…A language which allows this…

Subj > Obj > IO > OPrep > Poss > Subj > Obj > IO > OPrep > Poss > OCompOComp



A language which allows this…A language which allows this… Will also allow these.Will also allow these.




A language which allows this…A language which allows this… Will also allow these. But not these…Will also allow these. But not these…


Relation to L2A?Relation to L2A? Suppose that KoL includes where the Suppose that KoL includes where the

target language is on the NPAH.target language is on the NPAH. Do L2’ers learn the easy/unmarked/simple Do L2’ers learn the easy/unmarked/simple

relative clauses before the others?relative clauses before the others? Do L2’ers transfer the position of their L1 Do L2’ers transfer the position of their L1

first?first? Does a L2’ers interlanguage grammar Does a L2’ers interlanguage grammar

obey this typological generalization (if obey this typological generalization (if they can relativize a particular point on they can relativize a particular point on the NPAH, can they relativize everything the NPAH, can they relativize everything higher too?)?higher too?)?

NPAH and L2A?NPAH and L2A? Probably:Probably: The higher something is on the The higher something is on the

NPAH, the easier (faster) it is to learn.NPAH, the easier (faster) it is to learn. So, it might be easier to start by teaching So, it might be easier to start by teaching

subject relatives, then object, then indirect subject relatives, then object, then indirect object, etc. At each step, the difficulty would be object, etc. At each step, the difficulty would be low.low.

But, it might be more But, it might be more efficientefficient to teach the to teach the (hard) object of a comparison—because if (hard) object of a comparison—because if L2’ers interlanguage grammar includes L2’ers interlanguage grammar includes whatever the NPAH describes, knowing that whatever the NPAH describes, knowing that OCOMP is possible implies that everything OCOMP is possible implies that everything (higher) on the NPAH is possible too. That is, (higher) on the NPAH is possible too. That is, they might know it they might know it without instructionwithout instruction. (Same . (Same issue as before with the phonology)issue as before with the phonology)

NPAH in L2ANPAH in L2A

Very widely studied implicational Very widely studied implicational universal in L2A—many people have universal in L2A—many people have addressed the question of whether addressed the question of whether the IL obeys the NPAH and whether the IL obeys the NPAH and whether teaching aa marked structure can teaching aa marked structure can help.help.

Eckman et al. (1989) was about this Eckman et al. (1989) was about this second question…second question…

Change from pre- to Change from pre- to post-testpost-test

Eckman, Bell, & Nelson Eckman, Bell, & Nelson (1988)(1988)

Doughty (1991)Doughty (1991)

Investigating several issues at once:Investigating several issues at once: Effectiveness of type of instructionEffectiveness of type of instruction

Meaning orientedMeaning oriented Rule orientedRule oriented

Effectiveness of teaching “down the Effectiveness of teaching “down the markedness hierarchy” markedness hierarchy” (teaching a (teaching a marked structure and allowing marked structure and allowing learner-internal generalization to an learner-internal generalization to an unmarked structure).unmarked structure).


Subjects: Subjects: 20 international students 20 international students taking intensive ESL courses, taking intensive ESL courses, without much prior knowledge of without much prior knowledge of relative clauses. Average length of relative clauses. Average length of stay in the US was 3.7 months.stay in the US was 3.7 months.

Tasks:Tasks: Grammaticality judgmentGrammaticality judgment Sentence completionSentence completion


Subjects were pretested, then over two Subjects were pretested, then over two weeks (10 weekdays) they came in to a weeks (10 weekdays) they came in to a computer lab to take a “language lesson”. computer lab to take a “language lesson”. Then, immediately afterwards, subjects Then, immediately afterwards, subjects were posttested.were posttested.

In the language lessons, one of three In the language lessons, one of three possible things happened:possible things happened: Subject got the “meaning oriented treatment”Subject got the “meaning oriented treatment” Subject got the “rule oriented treatment”Subject got the “rule oriented treatment” Subject got the “control treatment”Subject got the “control treatment”


Daily lessons were a text of 5-6 Daily lessons were a text of 5-6 sentences (of a two-week long “story”) sentences (of a two-week long “story”) containing an relative clause formed containing an relative clause formed on the object of a preposition.on the object of a preposition. This is This is the book the book that I was looking that I was looking forfor..

Recall: Recall: Noun phrase accessibility Noun phrase accessibility hierarchyhierarchy::SU > DO > IO> SU > DO > IO> OPOP > GEN > > GEN > OCOMPOCOMP

Procedure…Procedure… Three steps:Three steps:

SkimSkim Reading for understanding (experimental Reading for understanding (experimental

section)section) ScanScan

Skim:Skim: Subjects saw the text for 30 Subjects saw the text for 30 seconds, with title, first sentence and last seconds, with title, first sentence and last sentence highlighted—this is to “get the sentence highlighted—this is to “get the idea” of what the text is about.idea” of what the text is about.

Procedure…Procedure… Reading for understanding: Reading for understanding: Each sentence Each sentence

displayed consecutively at the top of the displayed consecutively at the top of the screen. Three different possibilities:screen. Three different possibilities: MOG: MOG: Also saw dictionary help (2m) and Also saw dictionary help (2m) and

semantic explanations (referents, synonyms) semantic explanations (referents, synonyms) (2m), including relationship between head (2m), including relationship between head noun and relative pronoun.noun and relative pronoun.

ROG:ROG: Saw a little animated presentation of Saw a little animated presentation of deriving a OPREP sentence from two sentences deriving a OPREP sentence from two sentences (This is the book, I was looking for the book, (This is the book, I was looking for the book, This is the book which I was looking for)This is the book which I was looking for)

COG: COG: Saw each sentence, 2.5 minutes.Saw each sentence, 2.5 minutes.

Procedure…Procedure…

Scan. Scan. Re-scan paragraph in order to Re-scan paragraph in order to be able to answer two questions be able to answer two questions about it, then write out a summary about it, then write out a summary (NL).(NL).

PretestPretest

SS SUSU dodo IOIO OPOP GEGE OOCC

33 ++ ++ -- -- -- --

55 ++ -- -- ++ -- --

2121 ++ -- -- ++ -- --

77 ++ -- -- -- -- --

22 ++ -- -- -- -- --

66 ++ -- -- -- -- --

44 ++ -- -- -- -- --

11 -- -- -- -- -- --


1717 ++ ++ -- -- ++ --

2020 ++ -- -- -- -- --

1515 ++ -- -- -- -- --

1919 -- -- -- -- -- --

1414 -- -- -- -- -- --

1616 -- -- -- -- -- --


99 ++ ++ ++ ++ ++ --

88 ++ ++ -- ++ ++ --

1010 ++ -- -- -- -- --

1313 ++ -- -- -- -- --

1212 -- -- -- -- -- --

1111 -- -- -- -- -- --

MOGROG

CoG

PosttestPosttest


33 ++ ++ ++ ++ ++ ++

55 ++ ++ ++ ++ ++ ++

2121 ++ ++ ++ ++ ++ ++

77 ++ ++ ++ ++ ++ ++

22 ++ ++ ++ ++ -- --

66 ++ ++ ++ ++ -- --

44 ++ -- -- -- -- --

11 ++ ++ -- -- -- --


1717 ++ ++ ++ ++ ++ ++

2020 ++ ++ -- ++ ++ ++

1515 ++ ++ -- -- -- --

1919 ++ ++ ++ ++ -- --

1414 ++ -- -- -- -- --

1616 ++ -- -- -- -- --


99 ++ ++ ++ ++ ++ ++

88 ++ ++ ++ ++ ++ ++

1010 ++ ++ -- ++ -- --

1313 ++ -- -- -- ++ --

1212 ++ -- -- -- -- --

1111 -- -- -- -- -- --

MOGROG

CoG

Group mean gain scoresGroup mean gain scores

0

5

10

15

20

25

30

35

40

SU DO IO OP GEN OC

MOGROGCoG

ResultsResults

Both experimental groups showed strong positive Both experimental groups showed strong positive effectseffects (“Second Language Instruction Does Make (“Second Language Instruction Does Make a Difference”).a Difference”).

The control group did too (simply from exposure) The control group did too (simply from exposure) but not as dramatic.but not as dramatic.

Both types of instruction appear to be equally Both types of instruction appear to be equally effective with respect to gain in relativization effective with respect to gain in relativization abilityability..

Comprehension-wise, MOG scored 70.01 vs. ROG’s Comprehension-wise, MOG scored 70.01 vs. ROG’s 43.68 and CoG’s 40.64. Significant.43.68 and CoG’s 40.64. Significant.

Subjects improved basically following the NPAH Subjects improved basically following the NPAH by being taught just a marked position.by being taught just a marked position.

CommentsComments

Note that:Note that: ROG subjects improved in their ability ROG subjects improved in their ability

to relativize, yet didn’t do so well on the to relativize, yet didn’t do so well on the comprehension tests—comprehension tests—meaning isn’t meaning isn’t utmost in getting the structural rulesutmost in getting the structural rules..

MOG subjects got the structural MOG subjects got the structural properties even though not directly properties even though not directly instructed in them (instructed in them (meaning didn’t get meaning didn’t get in the wayin the way).).

What about markedness-What about markedness-based shortcuts?based shortcuts?

It looks like training them on OPREP It looks like training them on OPREP successfully brought subjects to be able to successfully brought subjects to be able to relativize on everything higher (Subj., Dir. relativize on everything higher (Subj., Dir. Obj., Indir. Obj.).Obj., Indir. Obj.).

But mysteriously, many people also seemed But mysteriously, many people also seemed to get OCOMP by the post-test.to get OCOMP by the post-test.

Interlanguage grammars Interlanguage grammars dodo seem to obey the seem to obey the typological requirements on languages typological requirements on languages (NPAH).(NPAH).

Is genitive mis-analyzed in the NPAH Is genitive mis-analyzed in the NPAH typological work, given that it seems to be typological work, given that it seems to be gotten early…?gotten early…?

Transfer, markedness, …Transfer, markedness, …

Do (2002) looked at the NPAH going Do (2002) looked at the NPAH going the other way, Englishthe other way, EnglishKorean.Korean. English: English: Relativizes on all 6 positions.Relativizes on all 6 positions. Korean:Korean: Relativizes on 5 (not OCOMP) Relativizes on 5 (not OCOMP)

Found a very similarFound a very similarpattern to what wepattern to what wesaw from Doughty’ssaw from Doughty’sexperiment.experiment.

SS SUSU dodo IOIO OPOP GEGE

1313 ++ ++ ++ ++ ++

1414 ++ ++ ++ ++ --

1616 ++ ++ ++ -- --

2929 ++ ++ -- -- --

3131 ++ -- -- -- --

2020 -- -- -- -- --

Transfer, markedness, …Transfer, markedness, …

The original question Do was looking at The original question Do was looking at was: was: Do English speakers transfer their Do English speakers transfer their position on the NPAH to the IL Korean?position on the NPAH to the IL Korean?

But look: But look: If English allows all 6 If English allows all 6 positions, why do some of the learners positions, why do some of the learners only relativize down to DO, some to IO, only relativize down to DO, some to IO, some to OPREP?some to OPREP?

We haven’t even reached the question We haven’t even reached the question of transfer yet—it looks like they start of transfer yet—it looks like they start over.over.

Subset principle?Subset principle?

Null subject parameterNull subject parameter Option (a): Null subjects are Option (a): Null subjects are permittedpermitted.. Option (b): Null subjects are Option (b): Null subjects are not permittednot permitted..

Italian Italian = option a, = option a, English English = option b.= option b.

E

I

A tempting analogy… in some cases, parameters seem to be ranked in terms of how permissive each setting is.

Reminder: Subset Reminder: Subset PrinciplePrinciple

The idea isThe idea is If one has only positive evidence, andIf one has only positive evidence, and If parameters are organized in terms of If parameters are organized in terms of

permissiveness,permissiveness, Then for a parameter setting to be Then for a parameter setting to be

learnable, the starting point needs to be learnable, the starting point needs to be the the subsetsubset setting of the parameter. setting of the parameter.

The Subset principle says that The Subset principle says that learners should start with the learners should start with the English setting of the null subject English setting of the null subject parameter and parameter and move to move to the Italian the Italian setting if evidence appears.setting if evidence appears.

E

I

Reminder: Subset Reminder: Subset PrinciplePrinciple

The The Subset PrincipleSubset Principle is basically that is basically that learners are learners are conservativeconservative—they only —they only assume a grammar sufficient to generate the assume a grammar sufficient to generate the sentences they hear, allowing positive sentences they hear, allowing positive evidence to serve to move them to a evidence to serve to move them to a different parameter setting.different parameter setting.

Applied to L2: Applied to L2: Given a choice, the L2’er Given a choice, the L2’er assumes a grammatical option that assumes a grammatical option that generates a subset of the what the generates a subset of the what the alternative generates.alternative generates.

Does this describe L2A?Does this describe L2A? Is Is thisthis a useful sense of markedness? a useful sense of markedness?

Subset principle and Subset principle and markednessmarkedness

Based on the Subset principle, we’d expect Based on the Subset principle, we’d expect the unmarked values (in a UG where the unmarked values (in a UG where languages are learnable) to be the ones languages are learnable) to be the ones which produce the “smallest” grammars.which produce the “smallest” grammars.

Given that in L1A we don’t seem to see any Given that in L1A we don’t seem to see any “misset” parameters, we have at least “misset” parameters, we have at least indirect evidence that the Subset principle is indirect evidence that the Subset principle is at work. Is there any evidence for it in L2A? at work. Is there any evidence for it in L2A? Do these NPAH results constitute such Do these NPAH results constitute such evidence?evidence?

Subset vs. TransferSubset vs. Transfer The Subset Principle, if it operating, would say The Subset Principle, if it operating, would say

that L2A starts with all of the defaults, the that L2A starts with all of the defaults, the maximally conservative grammar.maximally conservative grammar.

Another, Another, mutually exclusivemutually exclusive possibility possibility (parameter by parameter, anyway) is that L2A (parameter by parameter, anyway) is that L2A starts with the L1 setting.starts with the L1 setting. This means that for certain pairs of L1 and L2, This means that for certain pairs of L1 and L2,

where the L1 has the marked (superset) value and where the L1 has the marked (superset) value and L2 has the unmarked (subset) value, L2 has the unmarked (subset) value, only negative only negative evidenceevidence could move the L2’er to the right setting. could move the L2’er to the right setting.

Or, some mixture of the two in different areas.Or, some mixture of the two in different areas.

NPAH and processing?NPAH and processing? At least a plausible alternative to the NPAH results At least a plausible alternative to the NPAH results

following from the Subset Principle is just that following from the Subset Principle is just that relative clauses formed on positions lower in the relative clauses formed on positions lower in the hierarchy are hierarchy are harder to processharder to process. Consider:. Consider:

The astronaut…The astronaut… who who [[IPIP tt met me yesterday]met me yesterday] SUBSUB who who [[IPIP II [[VPVP met met tt yesterday]] yesterday]] DODO who who [[IPIP II [[VPVP gave a bookgave a book [[PPPP toto tt ]]]]]] IOIO who who [[IPIP I was I was [[VPVP talking talking [[PPPP about about tt ]]]]]] OPREPOPREP whose house whose house [[IPIP II [[VPVP likelike [[DPDP t t ’s house’s house]]]]]] GENGEN who who [[IPIP I am I am [[APAP bravebrave [[degPdegP -er-er [[thanPthanP than than tt ]]]]]]]] OCOMPOCOMP

NPAH and processing?NPAH and processing?

If it’s about processing, then the If it’s about processing, then the reason L2’ers progress through the reason L2’ers progress through the “hierarchy” might be that initially they “hierarchy” might be that initially they have limited processing room—they’re have limited processing room—they’re working too hardworking too hard at the L2 to be able at the L2 to be able to process such deep extractions.to process such deep extractions.

Why are they working so hard?Why are they working so hard? (Well, maybe L2A is like learning (Well, maybe L2A is like learning

calculus?)calculus?)

NPAH and processing?NPAH and processing? Is the NPAH Is the NPAH itselfitself simply a result of simply a result of

processing?processing? The NPAH is a typological generalization The NPAH is a typological generalization

about about languageslanguages not about the not about the course of course of acquisitionacquisition..

Does Arabic have a lower threshhold for Does Arabic have a lower threshhold for processing difficulty than English? Doubtful.processing difficulty than English? Doubtful.

The NPAH may still be real, still be a The NPAH may still be real, still be a markedness hierarchy based in something markedness hierarchy based in something grammatical, but it turns out to be grammatical, but it turns out to be confounded confounded by processing.by processing.

So finding evidence of NPAH position So finding evidence of NPAH position transfertransfer is very difficult. is very difficult.

Subset problems?Subset problems? One problem, though, is that many of the One problem, though, is that many of the

parameters of variation we think of today parameters of variation we think of today don’t seem to be really don’t seem to be really inin a subset- a subset-superset relation. So there has to be superset relation. So there has to be something else going on in these cases something else going on in these cases anyway.anyway. VVTT

Yes: √SVAO, *SAVOYes: √SVAO, *SAVO No: *SVAO, √SAVONo: *SVAO, √SAVO

Anaphor typeAnaphor type Monomorphemic: √LD, *Non-subjectMonomorphemic: √LD, *Non-subject Polymorphemic: *LD, √Non-subject Polymorphemic: *LD, √Non-subject

Mazurkewich (1984)Mazurkewich (1984) John gave a book John gave a book to Maryto Mary “unmarked”“unmarked” John gave John gave Mary Mary a book.a book. “marked”“marked”

To whom To whom did John give a book?did John give a book? “unmarked”“unmarked”?? WhoWho did John give a book did John give a book toto?? “marked”“marked”

AssumingAssuming that the second of each pair is marked, that the second of each pair is marked, Mazurkewich asked about timing of each in L2A.Mazurkewich asked about timing of each in L2A.

ButBut although maybe more languages allow the although maybe more languages allow the first of each pair than the second, the pied-piping first of each pair than the second, the pied-piping example should make us suspicious. Sounds kind example should make us suspicious. Sounds kind of stilted for being the unmarked option…of stilted for being the unmarked option…

Mazurkewich (1984)Mazurkewich (1984)

French-->EnglishFrench-->English and and Inuktitut--Inuktitut-->English>English

FrenchFrench lacks pied-piping and double- lacks pied-piping and double-object constructions.object constructions.

InuktitutInuktitut is different enough that it is is different enough that it is hard to find an analog to either the marked hard to find an analog to either the marked or unmarked constructions. or unmarked constructions. (or so it is (or so it is claimed)claimed)

Did the L2’ers prefer the unmarked Did the L2’ers prefer the unmarked structures? Did they acquire them first?structures? Did they acquire them first?

Mazurkewich (1984)Mazurkewich (1984)

French-L1 beginners do appear to “prefer” French-L1 beginners do appear to “prefer” the unmarked structures (2-to-1), and the the unmarked structures (2-to-1), and the marked structures gain ground as L2’ers marked structures gain ground as L2’ers become more advanced.become more advanced.

But French lacks the marked structure; did But French lacks the marked structure; did they “start with the unmarked structure” or they “start with the unmarked structure” or did they “start with the structure of their L1”?did they “start with the structure of their L1”?

As for Inuktitut, they As for Inuktitut, they weaklyweakly “preferred” the “preferred” the unmarked structures (beginners 77% to 98%).unmarked structures (beginners 77% to 98%).

Not very dramatic, not very convincing.Not very dramatic, not very convincing.

Mazurkewich (1984)Mazurkewich (1984) Worse, on a different task (“question the Worse, on a different task (“question the

italicized phrase”), although the French italicized phrase”), although the French speakers showed a moderate preference for speakers showed a moderate preference for “unmarked” (pied-piping) structures, “unmarked” (pied-piping) structures, the the Inuktitut speakers showed a preference Inuktitut speakers showed a preference for the for the markedmarked structure structure..

However, it could be that the whole experiment However, it could be that the whole experiment isn’t getting at what we want. The isn’t getting at what we want. The controlscontrols preferred the marked structure 3 or 4-to-1, so preferred the marked structure 3 or 4-to-1, so these “unmarked” structures seem to be these “unmarked” structures seem to be markedmarked from a language-internal perspective. from a language-internal perspective. Plus, this gives the learner a lot of evidence.Plus, this gives the learner a lot of evidence.

Problems so farProblems so far If L1 has an “unmarked” value for If L1 has an “unmarked” value for

something and L2 has a “marked” something and L2 has a “marked” value, if the L2’er prefers (or, better, value, if the L2’er prefers (or, better, learns more quickly) the “unmarked” learns more quickly) the “unmarked” value, it could be value, it could be eithereither transfertransfer oror reverting to an unmarked valuereverting to an unmarked value..

The actual marked/unmarked set The actual marked/unmarked set must be convincingly chosen—must be convincingly chosen—means nothing if we aren’t actually means nothing if we aren’t actually lookinglooking at marked/unmarked. at marked/unmarked.

Best test would be…Best test would be… Find a convincing marked vs. Find a convincing marked vs.

unmarked pair, …unmarked pair, … Find an L2 which allows only the Find an L2 which allows only the

marked option, …marked option, … Test speakers of an L1 which Test speakers of an L1 which alsoalso

only allows the marked option, …only allows the marked option, …

……and see if L2’ers use/accept the and see if L2’ers use/accept the unmarked option early on.unmarked option early on.

Liceras (1985, 1986)Liceras (1985, 1986)

Another potential marked/unmarked Another potential marked/unmarked pair:pair: Allows Ø comp. (marked; English)Allows Ø comp. (marked; English) Disallows Ø comp. (unmarked; Spanish)Disallows Ø comp. (unmarked; Spanish)

EnglishEnglishSpanishSpanish Beginners: 49% acceptance of Ø comp.Beginners: 49% acceptance of Ø comp. Intermediate: 25% acceptance.Intermediate: 25% acceptance. Advanced: 9% acceptance.Advanced: 9% acceptance.

Looks like Looks like transfertransfer (not initial (not initial unmarkedness) (contra Liceras’ unmarkedness) (contra Liceras’ hypothesis)hypothesis)

Schwartz (1993)Schwartz (1993) Back to the questions:Back to the questions:

How is a L2 acquired?How is a L2 acquired? Is L2 knowledge like native knowledge?Is L2 knowledge like native knowledge? Supposing it is, then Supposing it is, then knowing the rulesknowing the rules isn’t really part of isn’t really part of

knowing the language.knowing the language.

Of course, you can learn the rules and consciously Of course, you can learn the rules and consciously follow them. But is that knowing English?follow them. But is that knowing English? Prepositions are things you don’t end a sentence Prepositions are things you don’t end a sentence

with.with. Strive to not split your infinitives.Strive to not split your infinitives. Don’t be so immodest as to say Don’t be so immodest as to say I and John leftI and John left; say ; say

John and I left John and I left instead.instead. ImpactImpact is not a verb. is not a verb.

Schwartz (1993)Schwartz (1993)

Schwartz distinguishes two kinds of Schwartz distinguishes two kinds of knowledge:knowledge:

Learned linguistic knowledgeLearned linguistic knowledge I want to definitely avoid splitting my infinitives.I want to definitely avoid splitting my infinitives.

CompetenceCompetence *Who did John laugh after asking whether I *Who did John laugh after asking whether I

spread the rumor that bought the coffee?spread the rumor that bought the coffee?

L1AL1A UGUG (the range of possible (the range of possible

languages/grammars)languages/grammars) LADLAD (a system for getting from the data to (a system for getting from the data to

the particular parameter setting for the the particular parameter setting for the target language—not a conscious process, target language—not a conscious process, nor available to conscious introspection)nor available to conscious introspection)

PLDPLD (positive input) (positive input) Would it help the LAD to get rules explicitly?Would it help the LAD to get rules explicitly?

(“Use (“Use dodo to avoid stranding tense in Infl”; “Don’t to avoid stranding tense in Infl”; “Don’t extract an embedded subject out from under an extract an embedded subject out from under an overt complementizer”; “You want the other overt complementizer”; “You want the other spoon.”)spoon.”)

L2AL2A

If L1AD can’t really use this information, why If L1AD can’t really use this information, why would we necessarily think that the rules we would we necessarily think that the rules we learn in French class are in the right form to learn in French class are in the right form to “be absorbed” by the L2AD, if such a thing “be absorbed” by the L2AD, if such a thing exists…?exists…?

That is: L2 has things about it which can only That is: L2 has things about it which can only be learned with the help of negative evidence be learned with the help of negative evidence (or an L2AD). Yet this doesn’t guarantee that (or an L2AD). Yet this doesn’t guarantee that negative evidence negative evidence willwill help. help.

How can we How can we telltell the the difference between LLK difference between LLK

and competence?and competence? (Well-formulated) parameters have (Well-formulated) parameters have

wide-ranging effects. For example, wide-ranging effects. For example, verb raising:verb raising: *X: F question can’t use *X: F question can’t use dodo-support.-support. Y: F adverbs ok between V and Obj.Y: F adverbs ok between V and Obj.

Train subjects on *X. If they reset the Train subjects on *X. If they reset the parameter, parameter, a)a) they should they should “automatically” know Y as well, and “automatically” know Y as well, and b)b) they can use negative evidence. they can use negative evidence.

Schwartz’s modelSchwartz’s model

LAD KoL

blah blah blah

So why does it seem to So why does it seem to be useful to be taught be useful to be taught

the rules?the rules? Perhaps—knowing the rules (though Perhaps—knowing the rules (though

it is LLK) allows you in a way to it is LLK) allows you in a way to generate your own PLD. It’s that generate your own PLD. It’s that PLD, the output of using the rules, PLD, the output of using the rules, which the “L2AD” can make use of which the “L2AD” can make use of when constructing KoL.when constructing KoL.

This might explain the apparent This might explain the apparent truth that truth that practicingpracticing helps a lot more helps a lot more than just memorizing the rules…?than just memorizing the rules…?

Krashen’s “Monitor Krashen’s “Monitor Model”Model”

An early and influential model of An early and influential model of second language acquisition was the second language acquisition was the “Monitor Model”, based on five basic “Monitor Model”, based on five basic hypotheses:hypotheses: The Acquisition-Learning HypothesisThe Acquisition-Learning Hypothesis The Monitor HypothesisThe Monitor Hypothesis The Natural Order HypothesisThe Natural Order Hypothesis The Input HypothesisThe Input Hypothesis The Affective Filter HypothesisThe Affective Filter Hypothesis

The Acquisition-Learning The Acquisition-Learning HypothesisHypothesis

Acquisition and Learning are different.Acquisition and Learning are different. AcquisitionAcquisition refers to the (subconscious) internalizing refers to the (subconscious) internalizing

of implicit rules, the result of meaningful naturalistic of implicit rules, the result of meaningful naturalistic interaction using the language.interaction using the language.

LearningLearning refers to the conscious process that results refers to the conscious process that results in in knowing about knowing about the language, e.g., the result of the language, e.g., the result of classroom experience with explicit rules. classroom experience with explicit rules. (LLK)(LLK)

That is, you can learn without acquiring (or acquire That is, you can learn without acquiring (or acquire without learning).without learning).

Krashen hypothesizes that learned and acquired Krashen hypothesizes that learned and acquired rules are rules are storedstored differently; one cannot eventually be differently; one cannot eventually be converted into the other; they are simply different.converted into the other; they are simply different. Perhaps, or maybe the speculation on the previous slide was Perhaps, or maybe the speculation on the previous slide was

right.right.

The Natural Order The Natural Order HypothesisHypothesis

Acquisition proceeds in a “natural order” Acquisition proceeds in a “natural order” (i.e. the order of morpheme acquisition (i.e. the order of morpheme acquisition discussed earlier).discussed earlier).

This says nothing about learning, only This says nothing about learning, only acquisition.acquisition.

Also: Also: Krashen’s actual hypothesis is based on Krashen’s actual hypothesis is based on post-hoc analysis of the order L2’er post-hoc analysis of the order L2’er dodo seem to seem to acquire these morphemes—there’s no acquire these morphemes—there’s no underlying theoretical machinery. That’s not underlying theoretical machinery. That’s not to say that there to say that there couldn’tcouldn’t be some, of course. be some, of course.

The Monitor HypothesisThe Monitor Hypothesis

A linguistic expression originates in A linguistic expression originates in the system of acquired knowledge, the system of acquired knowledge, but prior to output a “Monitor” but prior to output a “Monitor” checks it against consciously known checks it against consciously known rules and may modify the expression rules and may modify the expression before it is uttered.before it is uttered.

Acquiredcompetence

Learnedcompetence(the Monitor)

output

The Monitor HypothesisThe Monitor Hypothesis For the Monitor to work, you need toFor the Monitor to work, you need to

Be able to focus on the form (time, attention)Be able to focus on the form (time, attention) Know the ruleKnow the rule

So, under pressure (e.g., time pressure), So, under pressure (e.g., time pressure), the Monitor may not be operating…the Monitor may not be operating…

Acquiredcompetence


output

The Monitor HypothesisThe Monitor Hypothesis

The Monitor would probably be the The Monitor would probably be the place where things like “don’t split place where things like “don’t split infinitives” and “don’t end a infinitives” and “don’t end a sentence with a preposition” live as sentence with a preposition” live as well.well.

Acquiredcompetence


output

The Input HypothesisThe Input Hypothesis

The Input Hypothesis draws on the The Input Hypothesis draws on the Natural Order Hypothesis; the idea is Natural Order Hypothesis; the idea is that there is a natural order of that there is a natural order of acquisition, but in order to advance acquisition, but in order to advance from one step to the next, a learner from one step to the next, a learner needs to get needs to get comprehensible inputcomprehensible input, , input which provides evidence for the input which provides evidence for the stage one level past the learners’ stage one level past the learners’ current level. The idea is that only this current level. The idea is that only this level of input is useful for the level of input is useful for the advancement of acquisition.advancement of acquisition.

The Input HypothesisThe Input Hypothesis Krashen’s view on acquisition: Speaking Krashen’s view on acquisition: Speaking

does not does not causecause acquisition, it is the result of acquisition, it is the result of acquisition, having built competence on the acquisition, having built competence on the basis of comprehensible input.basis of comprehensible input.

If input is at the right level and comes in If input is at the right level and comes in sufficient quantity, the necessary grammar sufficient quantity, the necessary grammar is automatically acquired.is automatically acquired.

The language teacher’s main role, then, is The language teacher’s main role, then, is to provide adequate amounts of to provide adequate amounts of comprehensible input for the language comprehensible input for the language learners.learners. Let’s stick to the model and not the politics Let’s stick to the model and not the politics

here…here…

Input ≠ intakeInput ≠ intake

InuktitutInuktitut——inputinput:: QasuiirsarvigssarsingitluinarnarpuqQasuiirsarvigssarsingitluinarnarpuq ‘‘Someone did not find a completely Someone did not find a completely

suitable resting place.’suitable resting place.’

tiredtired cause.because.be suitable suitable not not someonesomeoneQasu-Qasu-iiriir--sarsar--vigvig--ssarssar--sisi--ngitngit--luinarluinar--narnar--puqpuq notnot place.forplace.for find find completely completely 3sg3sg


After three long nights of After three long nights of gripninggripning, John , John finally found his finally found his slipwoggleslipwoggle..

Knowing so much about the rest of the Knowing so much about the rest of the sentence can tell us quite a bit about the sentence can tell us quite a bit about the parts we don’t know yet. (Slipwoggle is a parts we don’t know yet. (Slipwoggle is a noun, a possessible thing; to gripen(?) is a noun, a possessible thing; to gripen(?) is a verb, a process that one can perform over verb, a process that one can perform over an extended period of time). We can then an extended period of time). We can then make use of this to build our language make use of this to build our language knowledge (here, vocabulary).knowledge (here, vocabulary).


(Krashen)(Krashen) Learner must get Learner must get comprehensible inputcomprehensible input (mixture of (mixture of structures acquired and structures not structures acquired and structures not yet acquired) to advance.yet acquired) to advance.

Input: Input: What is available to the learner.What is available to the learner. Intake: Intake: Input that is Input that is used used in grammar-in grammar-

building.building.

What makes input into What makes input into intake?intake?

ApperceptionApperception: Recognizing the gap between : Recognizing the gap between what L2’er knows and what there is to know.what L2’er knows and what there is to know.

ComprehensibilityComprehensibility: Either the : Either the semanticsemantic meaning meaning is determinable or the relevant is determinable or the relevant structural aspectsstructural aspects are determinable.are determinable.

AttentionAttention: Selecting aspects of the knowledge to : Selecting aspects of the knowledge to be learned (from among many other possible be learned (from among many other possible things) for processing.things) for processing.

OutputOutput: Forcing a structural hypothesis, : Forcing a structural hypothesis, elsewhere used to shape input into a form useful elsewhere used to shape input into a form useful for intake.for intake.

Input Input apperception apperception

Some input is apperceived, some isn’t.Some input is apperceived, some isn’t. That which isn’t is thought of as That which isn’t is thought of as

blockedblocked by various “filters”: by various “filters”: Time pressureTime pressure Frequency non-extremesFrequency non-extremes Affective (status, motivation, attitude, …)Affective (status, motivation, attitude, …) Prior knowledge (grounding, analyzability)Prior knowledge (grounding, analyzability) Salience (drawing attention)Salience (drawing attention)

The Affective Filter The Affective Filter HypothesisHypothesis

Another aspect of the need for Another aspect of the need for comprehensible input is that it must comprehensible input is that it must be “let in” by the learner. Various be “let in” by the learner. Various “affective” factors like motivation, “affective” factors like motivation, anxiety, can “block” input and keep anxiety, can “block” input and keep it from effectively producing it from effectively producing acquisition.acquisition.

The overall modelThe overall model

Although Krashen’s “Monitor Model” Although Krashen’s “Monitor Model” suffers from a lack of specific suffers from a lack of specific testable details, it has had a testable details, it has had a significant impact on L2A research, significant impact on L2A research, and has an intuitive appeal.and has an intuitive appeal.

An interesting ideaAn interesting idea(courtesy of Carol Neidle)(courtesy of Carol Neidle)

If you were to learn French, you If you were to learn French, you would be taught conjugations of would be taught conjugations of regular and irregular verbs. Regular regular and irregular verbs. Regular --erer verbs have a pattern that looks verbs have a pattern that looks like this:like this: Infinitive: Infinitive: donnerdonner ‘give’‘give’ 1sg1sg je donnje donnee 1pl1pl nous donnnous donnonsons 2sg2sg tu donntu donneses 2pl2pl vous donnvous donnezez 3sg3sg il donnil donnee 3pl3pl ils donnils donnentent

Some French Some French “irregulars”“irregulars”

Infinitive: Infinitive: donnerdonner ‘give’‘give’ 1sg1sg je donnje donnee 1pl1pl nous donnnous donnonsons 2sg2sg tu donntu donneses 2pl2pl vous donnvous donnezez 3sg3sg il donnil donnee 3pl3pl ils donnils donnentent

Another class of verbs includingAnother class of verbs including acheteracheter ‘buy’ is classified as irregular, because the ‘buy’ is classified as irregular, because the vowel quality changes through the vowel quality changes through the paradigm.paradigm. Infinitive: Infinitive: ceder ceder ‘yield’‘yield’ 1sg1sg je cje cèèddee 1pl1pl nous cnous cééddonsons 2sg2sg tu ctu cèèddeses 2pl2pl vous cvous cééddezez 3sg3sg il cil cèèddee 3pl3pl ils cils cèèddentent


Infinitive: Infinitive: donnerdonner ‘give’‘give’ 1sg1sg je donnje donnee 1pl1pl nous donnnous donnonsons 2sg2sg tu donntu donneses 2pl2pl vous donnvous donnezez 3sg3sg il donnil donnee 3pl3pl ils donnils donnentent

The way it’s usually taught, you just have The way it’s usually taught, you just have to memorize that in the to memorize that in the nousnous and and vousvous form form you have “é” and in the others you have you have “é” and in the others you have “è”.“è”. Infinitive: Infinitive: ceder ceder ‘yield’‘yield’ 1sg1sg je cje cèèttee 1pl1pl nous cnous cééddonsons 2sg2sg tu ctu cèètteses 2pl2pl vous cvous cééddezez 3sg3sg il cil cèèttee 3pl3pl ils cils cèèddentent


However, the pattern makes perfect However, the pattern makes perfect phonologicalphonological sense in French—if you have a closed syllable sense in French—if you have a closed syllable (CVC), you get(CVC), you get èè,, otherwise you getotherwise you get éé..

[s[sd] (cède)d] (cède) [se.de] (cédez) [se.de] (cédez) So why is this considered irregular?So why is this considered irregular? Because in English, you think of the sounds inBecause in English, you think of the sounds in

cédezcédez asas [sed.de][sed.de], due to the rules of English , due to the rules of English phonology.phonology. Infinitive: Infinitive: ceder ceder ‘yield’‘yield’ 1sg1sg je cje cèèddee 1pl1pl nous cnous cééddonsons 2sg2sg tu ctu cèèddeses 2pl2pl vous cvous cééddezez 3sg3sg il cil cèèddee 3pl3pl ils cils cèèddentent


Because in English, you think of the sounds inBecause in English, you think of the sounds in cédezcédez asas [sed.de][sed.de], due to the rules of English , due to the rules of English phonology.phonology.

Since in all of these cases, Since in all of these cases, EnglishEnglish phonology phonology would have closed syllables, there’s no would have closed syllables, there’s no generalization to be drawn—sometimes closed generalization to be drawn—sometimes closed syllables have syllables have éé and sometimes they have and sometimes they have èè..

What could we do?What could we do? Infinitive: Infinitive: ceder ceder ‘yield’‘yield’ 1sg1sg je cje cèèddee [sed][sed] 1pl1pl nous cnous cééddonsons [sed.dõ][sed.dõ]

2sg2sg tu ctu cèèddes [sed]es [sed] 2pl2pl vous cvous cééddezez [sed.de][sed.de] 3sg3sg il cil cèèddee [sed][sed] 3pl3pl ils cils cèèddentent [sed][sed]


If people are really “built for language” and are If people are really “built for language” and are able to pick up language able to pick up language implicitlyimplicitly (as seems to be (as seems to be the case from everything we’ve been looking at), the case from everything we’ve been looking at), then if people are then if people are providedprovided with the right with the right linguistic data, they will more or less automatically linguistic data, they will more or less automatically learn the generalization.learn the generalization.

Problem is: The English Problem is: The English filterfilter on the French data on the French data is obscuring the pattern, and hiding the is obscuring the pattern, and hiding the generalization.generalization. Infinitive: Infinitive: ceder ceder ‘yield’‘yield’ 1sg1sg je cje cèèddee [sed][sed] 1pl1pl nous cnous cééddonsons [sed.dõ][sed.dõ]

2sg2sg tu ctu cèèddes [sed]es [sed] 2pl2pl vous cvous cééddezez [sed.de][sed.de] 3sg3sg il cil cèèddee [sed][sed] 3pl3pl ils cils cèèddentent [sed][sed]


Something to try: Something to try: Provide Provide people with the right people with the right data, see if they pick up the pronunciation. data, see if they pick up the pronunciation. Perhaps: Perhaps: exaggerate syllabificationexaggerate syllabification. (attention) . (attention) Perhaps try to instill this aspect of the phonology Perhaps try to instill this aspect of the phonology first.first.

Et voilàEt voilà. Perhaps this will make these “irregulars” . Perhaps this will make these “irregulars” as easy to learn as regularsas easy to learn as regulars!! The downside: I have no idea if this would actually work.The downside: I have no idea if this would actually work.

Infinitive: Infinitive: ceder ceder ‘yield’‘yield’ 1sg1sg je cje cèèddee “sed”“sed” 1pl1pl nous cnous cééddonsons “se—dõ”“se—dõ”

2sg2sg tu ctu cèèddes “sed”es “sed” 2pl2pl vous cvous cééddezez “se—de”“se—de” 3sg3sg il cil cèèddee “sed”“sed” 3pl3pl ils cils cèèddentent “sed”“sed”

““Incomprehensible Incomprehensible input”input”

So this is another way in which input So this is another way in which input might be “incomprehensible”—not might be “incomprehensible”—not that it is inherently that it is inherently incomprehensible (i.e. not that it incomprehensible (i.e. not that it would be incomprehensible to a would be incomprehensible to a L1’er), but that the L1’er), but that the prism of the L1prism of the L1 is getting in the way of seeing the is getting in the way of seeing the data for what it really is.data for what it really is.

Some critiques on record Some critiques on record re: the Monitor Modelre: the Monitor Model

Are acquired and learned rules really stored so Are acquired and learned rules really stored so separately that they cannot interact? Gass & separately that they cannot interact? Gass & Selinker’s textbook points out that “Selinker’s textbook points out that “it is it is counterintuitive to hypothesize that nothing learned counterintuitive to hypothesize that nothing learned in a formal situation can be a candidate for [fluent, in a formal situation can be a candidate for [fluent, unconscious speech]unconscious speech]”.”.

But this doesn’t seem to be a very persuasive But this doesn’t seem to be a very persuasive objection—First, objection—First, counterintuitiveness is not an counterintuitiveness is not an argumentargument. Second, even if formal, learned rules are . Second, even if formal, learned rules are stored completely separately, nothing prevents the stored completely separately, nothing prevents the useuse of these rules in production from providing of these rules in production from providing input input to the acquisition system, providing an to the acquisition system, providing an indirect “conversion” of knowledge.indirect “conversion” of knowledge.


G&S also observe (attributing the objection to G&S also observe (attributing the objection to Gregg) that in Krashen’s model, the Monitor only Gregg) that in Krashen’s model, the Monitor only affects output (speech, writing), but anecdotal affects output (speech, writing), but anecdotal evidence for use of formally learned rules in evidence for use of formally learned rules in decoding heard utterances is easy to come by.decoding heard utterances is easy to come by.

Perhaps this is true of Krashen’s particular Perhaps this is true of Krashen’s particular statement, but there seems to be no need to toss out statement, but there seems to be no need to toss out all aspects of his hypotheses based on an oversight all aspects of his hypotheses based on an oversight of this sort—it seems easily repairable by extending of this sort—it seems easily repairable by extending the model to allow learned competence to also the model to allow learned competence to also monitor input and provide input to the acquired monitor input and provide input to the acquired competence.competence. Of course, Krashen may have Of course, Krashen may have meantmeant it, but that’s irrelevant. it, but that’s irrelevant.

He’s one guy with good ideas and bad ideas like anyone.He’s one guy with good ideas and bad ideas like anyone.


Most of the objections to the Monitor Model Most of the objections to the Monitor Model focus on the impreciseness of the hypotheses; focus on the impreciseness of the hypotheses; although Krashen may not have treated them this although Krashen may not have treated them this way, they clearly must be used only as a starting way, they clearly must be used only as a starting point, a way to think about the process of L2A.point, a way to think about the process of L2A.

Further research in this direction needs to be Further research in this direction needs to be focused on trying to refine the existing focused on trying to refine the existing “hypotheses” to yield testable (falsifiable) “hypotheses” to yield testable (falsifiable) hypotheses with a higher degree of specificity.hypotheses with a higher degree of specificity.

week 12. language universals, and the beginnings of a model grs lx 700 language acquisition and...

Documents