New User Special Price Expires in

Let's log you in.

Sign in with Facebook


Don't have a StudySoup account? Create one here!


Create a StudySoup account

Be part of our community, it's free to join!

Sign up with Facebook


Create your account
By creating an account you agree to StudySoup's terms and conditions and privacy policy

Already have a StudySoup account? Login here

Introduction to Linguistics

by: Hermina Hoppe

Introduction to Linguistics LIN 401

Marketplace > Michigan State University > Linguistics > LIN 401 > Introduction to Linguistics
Hermina Hoppe
GPA 3.77


Almost Ready


These notes were just uploaded, and will be ready to view shortly.

Purchase these notes here, or revisit this page.

Either way, we'll remind you when they're ready :)

Preview These Notes for FREE

Get a free preview of these Notes, just enter your email below.

Unlock Preview
Unlock Preview

Preview these materials now for free

Why put in your email? Get access to more of this material and other relevant free materials for your school

View Preview

About this Document

Class Notes
25 ?




Popular in Course

Popular in Linguistics

This 73 page Class Notes was uploaded by Hermina Hoppe on Saturday September 19, 2015. The Class Notes belongs to LIN 401 at Michigan State University taught by Staff in Fall. Since its upload, it has received 22 views. For similar materials see /class/207376/lin-401-michigan-state-university in Linguistics at Michigan State University.


Reviews for Introduction to Linguistics


Report this Material


What is Karma?


Karma is the currency of StudySoup.

You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!

Date Created: 09/19/15
Linguistics 401 section 3 Syntax Adjuncts Relative Clauses amp Coordination OCtOber 23 2007 1 Adjuncts Last weekls lectures used patterns of acceptability judgments to argue that words impose subcat egorization requirements on their complementsi Consider slay in its transitive frame NP i In an example like 1 the subcategorization requirement is met by the complement NP the dragon 1 Saint George slew Np the dragon with a sword But what can be said about the PP with a sword This phrase is not required by the subcategoriza tion frame of slagi Rather it is an adjunctlr 7 of which there can be many Example 2 has just ve PP adjuncts but as anyone who has ever played Clue can tell you any limitation on the number of possible adjuncts has more to do with the subject under discussion than with the grammar of the language 2 St George slew the dragon ppwith a sword ppon Tuesday pp in the parlour ppwith the maid ppwhile rakishly smoking a cigarillo 11 Doso substitution One can diagnose whether or not a phrase is a verbal adjunct using doso substitutioni If a phrase need not be included as part of the sequence being replaced by do so then it is an adjunct Otherwise if it must be included to preserve acceptability then it is a complement Example 3 shows a positive result from this test indicating that out of love is an adjunct 3 at Lancelot pursues Guinevere out of love and Arthur does so too bi Lancelot pursues Guinevere out of love and Arthur does so out of political necessityi Example 4 shows a negative result from this test indicating that the proper name Arthur is not serving as an adjunct This is consistent with the N P subcategorization frame for giver 4 at The Lady of the Lake gives Arthur his sword and Merlin does so too bi The Lady of the Lake gives Arthur his sword and Merlin does so the round table 12 Adjectival adjuncts There can be any number of prenominal adjectives in English They share this free iterability property with postverbal PP adjunctsi 1 2 1 2 3 4 5 5 The little red lights not blinking on my big black plastic Japanese cordless phone lAdjuncts often function as modi ers and the O7Grady books online supplement proceeds from this terminology This week we will come to know them syntactically as adjuncts deferring discussion of their meaning contribution until the unit on semantics 13 Adjunctions in Xbar theory Lets extend the X schema to handle adjunctsi They can come either on the left of a head eigi adjectives before a noun or on the right of a head eigi PP modi ers In at least the verbal case they come after any subcategorized complements To capture these properties in our theory of phrase structure introduce a new structurebuilding rule that merges in an adjunct at the X level The mother node dominating these two children will have the same category as the nonadjunct daught er XP XP S X Speci er X p601 er Adjunct X X Adjunct X Complement X Complement a Adjunct precedes the head b Adjunct follows the head Figure l X schema extended to handle adjuncts The extended X schema in gure 1 notates the complementadjunct distinction using the tree structural difference between SISTER OF X and SISTER OF Xi For instance in example 6 the rst PP of poems is a complement of the N book whereas the second PP with the glossy cover is attached as an adjunct 6 De P N l the N PP 1AA PP with the glossy cover N t book of poems Note well that book of poems with the glossy cover is an N the same syntactic category as the smaller Nbook of poems The merge rules that permit such an N to embed a smaller N exhibit recursion on the symbol N i All tree branches must instantiate one of these ruletypes headcomplement X A X Z where VVVXYZ range over syntactic categories the Xbarspeci er XP A Y X comma abstracts over linear ordering and parentheses in adjunctl A X adjunct2 XP A XP V dicates optionalityi Figure 2 Concise statement of twolevel X bar theory The head X is sometimes referred to as an X0 because it constitutes the zero h bar leveli X X1 is the rst bar level X2 is known as XP the maximal projection77 of the head 2 Relativization Relative clauses can be analyzed with a transformational movement rule not entirely unlike the one for information questionsi Movement of the Wh word is motivated by a relativization feature on C input NP output NP Det N Det N I the the N CP N CP 0 1 NP 0 I sorceress sorceress CR8l 1P 1I CHM H N NP 1 NP 1 A who Merlin Ipast VP A Merlin Ipast VP v VNP VNP taulght ll taulght if Ill who Figure 3 Relativization transformation The idea that the landing site is the speci er of CP is supported by examples where both C and its speci er are lled English7 at least nowadays7 appears not to allow this option 7 Dutch lk weet niet wie of Jan gezier heeft I know not whom whether Jan seen has I don7t know whom Jan has seen77 8 Bavarian German I woass ned wann dass da Xavea kummt I know not when that Xavea comes I don7t know when Xavea is coming77 Relativization can target any grammatical relationi When launched from subject position7 the movement does not actually cross any overt words and is said to be stringvacuousi 9 a the girl who 75 got the right answer is clever b the dog which Penny bought 75 today is very gentle c the man who Stephen explained the accident to t is kind d the ship which my uncle took Joe on t was interesting 3 Coordination Lets put words like and and or in a syntactic category Con and observe three properties 1 any category X X or XP may be coordinated by Con 2 the coordinated elements are typically share the same syntactic category 3 the conjoined result typically has the same category as the conjoined phrases Property 1 is demonstrated for maximal projections of NVAP by the coordination test examples from October ll h repeated here as 11 and for intermediate projections V A and P in 12 11 a Np Jane s shoes and Np the books in the living room take up too much space bi During lecture Mark Vp got bored and Vp started playing with his Treo c The really Ap depressed and Ap angry postdoc decided to jump off a bridge di They chased the getaway Car Pp around the corner and pp down the alley 12 a Sir Gawain rarely vgoes out and vgets wasted Morgan Le Fay favors the very Abright blue and Adull green gowni Arthur chased Sir Lancelot right pout of Britain and pinto France 0quot n One might begin to analyze these constructions using a at coordination schema as in OlGrady p184 gure 527 But doing so would force yet another extension of Xbar theoryi lnstead lets assume that the second conjunct attaches to the rst by XPadjunction as in 13 where V ConPi IP NP 1 NP ColnP uncover the betrayal Agravain 3011 Con NP 13 and Mordred The examples in 14 are consistent with the second conjunct s being a headinitial ConP rather than the rst conjunct forming a constituent with Con at its rightmost edger That grouping is not independently moveable the way ConP seems to be 14 a John bought a book and a newspaper yesterdayi bi John bought a book yesterday and a newspaper cl 96 John bought a newspaper yesterday a book and This Adjoined Con77 approach Munn 1993 is compatible with the occasional exception to Prop erty 2 where unlike categories are acceptably coordinatedi For instance in example 15 the copula is has two different transitive subcategorization frames and Coordinating those two in either order results in an acceptable complement 15 a Pat is either asleep or at the of ce bi Pat is either at the of ce or asleep Linguistics 401 section 2 Hale October 29 2003 Syntax addons to phrase structure grammar 1 Naive CFGS miss case phenomena 1 a gt1lt her will suffocate him in the forest b she will suffocate him in the forest I a case of overgeneration I 1167 8116 both category Pronoun I DP gt Pronoun any pronoun can occupy any DP position is not quite true Only the nominative case pronoun is acceptable in subject position 1 and only the accusative case pronoun is acceptable in object position 2 2 a she will suffocate he in the forest b she will suffocate him in the forest Case can be viewed assigned to DPs in much the same was theta roles Only some times is it morphologically marked ie quotndI quot 2ndP nominative accusative genitive me my you you your Specifier of T bears nominative case ho him his Complement of V bears accusative case she her her we us our they them their case lter only trees satisfying these rules are generated more specifically the case of subject DPs depends on the finiteness of the enclosing clause 3 a The chair indicated quot nhe INOM should teach LIN401 next semester b The chair wanted TPEFIHHQ meACC to teach LIN401 next semester analyze this alternation a syntactic dependency between T and T s specifier 2 Dependency 1 r 1 J 39 L h exists between positions p and q if replacement of the piquot word requires a corresponding replacement of the 11quot word to preserve language membership Example number agreement eat selects a plural but not singular subject 4 a the rats eat the cheese b rat eat the cheese there is an agreement dependency between the 5th word ratrats and the 10th word eateats 5 a the cheese that the rats that the cat saw eat stinks b the cheese that the rat that the cat saw eat stinks rats gt rat c the cheese that the rat that the cat saw eats stinks eat gt eats 3 Longdistance dependencies There is a dependency between which child and the underlined spot in 6 6 Which child do you think i ate all the cookies Replacement of the underlined blank with some child requires deletion ie replacement with nothing of the fronted witphrase 7 a Which child do you think some child ate all the cookies b Do you think some child ate all the cookies Likewise s selection Which planet do you think ate all the cookies planets donquott eat cookies c selection Which VP getting happy do you think ate all the cookies VP not accept able complement agreement in the present tense verb morphology shows number agreement gt1lt Which children do you think eats all the cookies plural singular gt1lt Which child do you think eat all the cookies singular plural The dependency can hold of arbitrarilylarge phrases 8 a Which child of strong cookieeating stock do you think ate all the cookies b Which child of strong cookieeating stock and opportunistic bent do you think ate all the cookies c Which child of strong cookieeating stock sizable tummy and opportunistic bent do you think ate all the cookies and can span arbitarily long distances a Which child does the father think ate all the cookies b Which child does the somewhat confused man in the plaid jacket think ate all the cookies c Which child does the somewhat confused man in the plaid jacket fumbling around in an empty cookie jar think ate all the cookies 4 Transformations account for longdistance dependencies Generate deep structure Twp using CFC like 7b H check to make sure case scselection and agreement wh movement requirements are satis ed 5 create a new surface structure 7um where comple ment witDP is now in specifier of JP 93 771mm 2 T VP l PRESENT V 1P2 l think DP VP Whidl Child ate all the cookies 1P1 DP 1 A which child Tom pm 2 1 TP l do DP TI l you T VP l PRESENT V 1P2 l l k t nn VP ate all the cookies shorthand lowercase t for tracequot of movement Which child do you think t ate all the cookies Method acceptability judgment Easy vs difficult sentences centerembedding that are all grammatical Easy vs difficult sentences centerembedding that are all grammatical a the reporter disliked the editor Easy vs difficult sentences centerembedding that are all grammatical a the reporter disliked the editor b the reporter who the senator attacked disliked the editor Easy vs difficult sentences centerembedding that are all grammatical a the reporter disliked the editor b the reporter who the senator attacked disliked the editor cthe reporter who the senator who John met attacked disliked the editor Easy vs difficult sentences centerembedding that are all grammatical 1 English center embedding with RCs a The patient the nurse likes was discharged yesterday b The patient the nurse the surgeon trusted liked was discharged yesterday 2 English right branching with CP Verbs a The nurse thought that the patient had been discharged b The nurse thought that the patient believed that the surgeon was incompetent Easy vs difficult sentences centerembedding that are all grammatical 3 Korean GP taking a na num Chelsvvu ka wulki sicakhayss tako mitnunta I Top Chelsvvu Nom cry started that believe I believe that Chelsvvu started to cry Sunhi nun Chelswu ka Yenghi ka uless ako malhayss tako mitnunta Sunhi Top Chelswu Nom Yenghi Nom cried that said that believes Sunhi believes that Chelsvvu said that Yenghi cried c Minca nun Sunhi ka Chelswu ka Yenghi ka uless tako malhayss tako sayngkakhan tako mitnun ta Minca Top Sunhi Nom Chelswu Nom Yengi Nom cried that said that thinks that believes Minca believes that Sunhi thinks that Chelsvvu said that Yenghi cried Top Topic marker Nom Nominative case marker Method selfpaced reading Easy vs difficult sentences SUbieCt VS Object that are all grammatical 39eXtraCted relatives Easy vs difficult sentences SUbieCt VS Object that are all grammatical 39eXtraCted relatlves 1the reporter who t sent the photographer to the editor hoped for a story Easy vs difficult sentences SUbieCt VS Object extracted relatives that are all grammatical 1the reporter who t sent the photographer to the editor hoped for a story 2the reporter who the photographer sent t to the editor hoped for a story Selfpaced reading Selfpaced reading Selfpaced reading reporter Selfpaced reading Selfpaced reading Selfpaced reading Selfpaced reading photographer Selfpaced reading Selfpaced reading Selfpaced reading Selfpaced reading Selfpaced reading Selfpaced reading Selfpaced reading subject relative thereportewho sent rmtograpfter the editohoped for a goodstory object relative thereportewho Wmtograpslmlrt to the editohoped for a goodstory subject vs objectextracted relatives subject relative thereportewho sent p rwtographer the editohoped for a goodstory object relative thereportewho Wmtogrammm to the editohoped for a goodstory subject vs objectextracted relatives a bump when subcat frame of send is saturated subject relative thereportewho sent ntwtographer the editohoped for thereportewho ntwtograptmm to the editohoped for a good story a good story subject vs objectextracted relatives a bump when suboat frame of send is saturated a bump on embedded Vsend 5 in Objextracted only subject relative thereportewho sent ntwtographer the editohoped for a good story thereportewho ntwtograptmm to the editohoped for a good story Easy vs difficult sentences SUbieCt VS Object that are all grammatical 39eXtraCted relatives She 2 1the repmhe photographer to the editor hoped for a story on3 1 2the reporter who the photographer to the editor hoped for a story subject vs objectextracted relatives interaction with memory ability Subject Relative Object Relative LovvU dim High MEAN READING TIME PER WORD mseclwordi The reporter senator admitted the The reporter attacked admitted the that error that error attacked the the senator Method eyetracking Easy vs difficult sentences centerembedding that are all grammatical 4 Japanese center embedding with RCs a Hiroshi ga Masao ga katta pan o tabeta Hiroshi Norn Masao Norn bought bread Ace ate Hiroshi ate the bread Masao bought b Yoko ga Hirorni ga Asako ga kaita genkoo o kakinaoshita syorui o yonda Yoko Norn Hirorni Norn Asako Norn wrote draft Ace rewrote paper ACC read Yoko read the papers that Hirorni rewrote based on the draft Asako wrote Nom Nominative case marker ACC Accusative case marker eyetracking Easy vs difficult sentences 2 that are all grammatical centerembedding W1 in Japanese 4 Japanese center embedding with RCS a t roshrga hdaaxyga ka x panaj tabeta Hiroshi Nom Masao Nom bought bread Ace ate t roshiatet uabreadhdasao39bought Reading Time msec CEDegraa 1 LBDegree 1 Simple V 6 Kanji Character Ea Sentence Type Thus one Kanji may correspond to many moras In addition to Kanji Japanese uses two types of syllabic alphabets called Hiragana and Katakana Each kana character in these syllabaries represents one mora Ordinarily Japanese sentences employ a mix of these three types of characters Kanji is used for content words most often and Hiragana is used for particles and inflections In print each character takes up the same amount of Fg 4 Read39ng Tme per Mora for Center space and each mora takes approximately the same length of time 7 j j 53 when it is uttered Thus the written length of a sentence does not exactly correspond to its length when read aloud depending on e entences how many Kanjis are used and how many mores each Kanji represents Method eventrelated potentials Avnraninn A Individual subjects B Grand average Nl2 r r II 6 f 39 t 39 I a z39 l 0 9 quot 7 9 V ltquot quot3f vquot 9 p v fi v mSGC Fig 2 Difference waves from experiment 2 strong semantic incongruity These di erence waveforms were obtained by subtracting the averaged ERP s to the semantically congruous words from the ERP s to the semantically incongruous seventh words Each superimposed tracing A represents the difference wave from one subject The ERP s in B are the corre sponding grand average waveforms over all 12 subjects 39 eventrelated potentials Semantic Deviation Bobcats hunt mice squirrels rabbits laughs and many other small N400 Grammatical Deviation Turtles will spit out things they does not like to eat P600 Linguistics 401 section 2 Hale October 15 2003 Phonology allophones amp phonemes Phonemic analysis a theory of distinction Observation to be explained Some of the possible phonetic distinctions in a language are systematically related to their surrounding context For instance not all s are the same in English subscript ering means voiceless leaf blue bluw plow plaw gleam Jlijm clap klaep 1 slip shp clear klij r og flag play plej l 2 What s the distribution after ivoice stop no elsewhere There is a surprising amount of this kind of structure in the sound patterns of natural languages Why coarticulation overlapping articulation of segments that are adjacent to one another This overlap happens because transitions between vocal tract states are gradual phonological processes a change in articulation conditioned by nearby sounds Although many processes can be understood as originating from coarticulation there exist others that are purely grammatical Different languages have different processes 3 Example nasal assimilation in English superscript tilde means nasalized oEnglish vowels are inasal in most contexts bit into an apple lis was reviewed by a lawyer bug where peat moss grows ovowels becomes nasal when followed by a nasal consonant bin threw into a lin against a wall bag the sound a bell makes Every vowel phoneme has in fact two allophones m zr big x17 cattle 4 Scots Gaelic has a similar process 2 means long vowel n zl cloud mu about rum secret Vowels are nasal in Scots Gaelic when preceded or followed by a nasal consonant m wah luxurious xnajag stalk marah scold na ascend 5 Malay doesl too fm lalrag forbid makan eat rumah house ksreta car All vowels and glides following a nasal in Malay are predictably nasalized until a non nasal consonant is reached For the French nasality is contrastive me but m hand page 545 in F romkin me very tu train Explanation Postulate two levels 1 underlying form string of phonemes that are distinctive in this language 2 surface or phonetic form specifying particular phones in a language independent way and provide phonological rules that derive the surface form from the underlying form The rule format we will use is A a B 07D A rewritten as B in the context where it is preceeded by C and followed by D77 feature Any of A B C D might be written more generally as a feature matrix feature A feature matrix makes positive or negative reference to a combination of phonological features like nasal voiced high affricate etc Such combinations specify natural classes For instance further English voicing data indicates that r patterns with l just as in 1 bruw prow grijn trip 7 drip dnp creep frag pray l so do and w beauty bj uwrij putrid pluwtpd Duane dwejn twin m m 8 Gwen g wan quick k view vjuw cute k juwt swim swmi thwack ewaek i English approximants have voiceless approx1mant i stop i i a 7 v01ce i 7 allophones after v01celess stops else v01ce 7 v01ce i i where such approx1mants are v01ced Linguistics 401 section 2 Hale October 21 2003 Syntax phrase structure 1 Containment is represented by dominance Phrases are sequences of words the things that informally receive thematic roles and are to a first approximation intersubstitutable LOCATION 1 The vicepresident resides pp in a secure location at the White House beyond Al Qaeda s grasp 2 The vicepresident resides Wide wasmngmn D c near Wis Crawford TX ranch in addition phrases can contain other phrases For instance the prepositional phrase in contains a determiner phrase a subconstituent 3 pp in DP a secure location Containment of DP within PP represented by the parentchild relationship in a tree PP P DP in a secure location where DP is a child of PP We that in this diagram PP dominates DP A more detailed phrase structure for 1 is given below L DP VP D N V PP l The vicepresident resides P DP l in y D Adj N l l l a secure location The labelledbracketing 5 0p 1 The I vice president I I W V resides pp p in I up 1 a M3 secure location I I I I I is a completelyequivalent notational variant of the phrase structure diagram 11 Terminology roundup node a place in a tree distinguished places include root a node that the child of no other has no siblings 8 is an ancestor to all leaf a node that has no children internal node neither root nor leaf label decoration written at a node in syntax typically the category of the constituent VP PP DR branching refers to number of lines emanating downward non has 1 child binary has 2 children ternary has 3 children sisters nodes are children of the same parent dominates contains within X dominates Y just in case there is a purely downward path from X to Y precedes comes before X precedes Y just in case some dominator of X and some domi nator of Y are sisters immediate dominance the single nextbigger constituent X is the shortest constituent longer than Y that contains Y immediate precedence comes right before there exists no Z such that X precedes Z and Z precedes Y 4 A U Bl In tree 4 all of these statements are true A dominates BCD and E B and C are children of A C dominates D and E D and E are children of C A immediately dominates B and C B and C are sisters 0 immediately dominates D and E D and E are sisters Linguistics 401 section 2 Hale November 4 2003 Semantics set theory and truth conditions 1 Truthconditional semantics realworld knowledge truth such not part of semantics which pieces of real world knowledge bear on sentencesquot truth the very stuff of semantics Meaning relationships Entailment 2 a Mary was laughing and dancing b Mary was dancing 3 a Mary is a great dancer b Mary is a dancer If 81 entails 82 its impossible for 81 to be true and 82 to be false in any situation Synonymy 4 a John sold a car to Mary b Mary bought a car from John 5 a John is in front of Mary b Mary is behind John Synonymous sentences are true in exactly the same situations mutually entailing Contradiction 6 a John is asleep b John is awake 7 a John claimed that semantics is the funnest part of linguistics b John denied that semantics is the funnest part of linguistics Contradictory sentences cannot both be true in the same situation Compositionality 8 a The horse behind Pegasus is bald true if T01 b The horse behind the horse behind Pegasus is bald true if T02 c The horse behind the horse behind the horse behind Pegasus is bald true if T03 The truth conditions of 8c are systematically related to the truth conditions of 8b compositionality The meaning of an expression is determined by the meaning of its parts and the way in which they are combined G Frege 1892 1 Modeltheoretic semantics in two pages I standardize on completely unambiguous picturequot of situations a model 39 1 fsetsof Pormalize real or imaginary entities in a situation individuals formalize properties sets of individuals An assignment function links particular words to particular entities in a model Compositional rules specifying how the meanings of children combine to yield the meanings of their parents permit the derivation of truth conditions for arbitrarilylong sentences 11 Set theory The things collected together in sets are the members of the set Members can be anything the set of Beatles E John Paul George Ringo the set of presidents of the USA in 2003 P2003 George W Bush colors in crayola 4pack C red green yellow blue the positive integers less than 7 87 1 2 3 4 5 6 7 the positive integers greater than 7 7 7 8 9 the set of English sentences Lngnsh Hi there Do you come here often 3 empy set E 4 3 0 the number of members of a set is its cardinalit l y P2003l 1 Cl El 4 r KIN f M l 39quot L The thing about a set is you re either in it or not Mka 9 E 10038360 My 10 m g quot1991 If one set has all the members of another unequal set they are in the proper subset relation 9 a 1 b C 5 baegic b 1 bj g 5 b a 6 110 Operations are usually defined on pairs of sets A B union anything in either set A U B E A or a E B intersection members common to both sets A O B E A and a E B difference members in one but not the other A B E A and a 6 B Compositional rules gives the truth conditions of the S in terms of the meaning of the DP and VP 10 a When DP is a proper name a sentence of the form DP VP is true if and only if DP is a member of VPE b DP Name Name 9 l W V l W Linguistics 401 section 2 Hale October 22 2003 Syntax contextfree grammar and Xbar theory 1 Contextfree phrase structure grammar Tl ees can depict the phrase structure of a sentence the constituents their categories and containment relationships These claims about sentences in a language can be verified with constituency tests such the Coordination Test Topicalization Test Fragment Test etc Contextfree phrase structure grammar is a theory of this structure It defines the set of phrase structures that are possible using the formalism of context free rewrite rules 1 contextsensitive rewrite rules conditions come after the slash continuant gt s read gl irate svllableinitiallv VOW l r a a i r voiced consonantal gt long i quotlengthen vowels before voiced consonantsquot consonantal A gt B C i D rewrite A B just in case A is preceded by C and followed by D 2 contextfree rewrite rules no slash no conditions PP gt P DP PP can immediately dominate P and DP in that order CP gt C 8 JP can be the parent of sisters 0 and 8quot VP gt V DP PP if you have a V a DP and a PP next to one another then you can have a VP containing them all Each such rule defines a little bit of phrase structure PP CP VP P DP C S V DP PP Contextfreeness means that any parent with a matching label can extend the tree inde pendent of existing sisters A very small theory of English syntax JF rewrite rules lexicon 1 S gt DP VP D the 2 DP gt D N N binoculars spy cop bridge 3 DP gt DP PP P with on 4 PP gt P DP V saw 5 VP gt V DP 6 VP gt V DP PP A derivation is a sequence of rewrite rule choices that culminate in the grammatical cat egories of a given sentence It always begins with the startquot category typically 8 If on a particular grammar a derivation exists for some sentence we that grammar generates the sentence IPhrase ructure has eff on human performance too People tend to make more mistakes remem bering individual words of sentences at major phrase boundaries Johnson 1965 as well as more mistakes understanding spoken words obscured by white noise Levelt 1970 2 ambiguity Can our Very Small Theory account for sentence 3 539 3 the spy saw the cop with the binoculars The syntax affects the meaning just in arithmetic 5 10 3 9 77 value of expression N semantics r f r r N V i depends on grouping a syntax Does our Very Small Theory generate sentence 4 4 the cop saw with the binoculars maria heard with the parabolic microphone How many adjunctPPs can we have 5 the cop with the binoculars saw the spy the cop with the binoculars on the bridge saw the spy the cop with the binoculars on the bridge near the pier by the saw the spy Analyze the infinite productivity of language recursion 6 Example DP recursion DP DP DP DP DP my best friend s sister s boyfriend s brother s girlfriend heard from this guy who knows this kid who s going with a girl who saw Ferris pass out at 31 flavors last night What rule do we need to handle 7 7 I think that she thinks that he thinks that she thinks that he likes her 3 X bar theory Q What s common to all these rules AP gt A PP PP gt P DP VP gt V DP DP gt D NP A The have the general form XP gt X ZP XP is the maximal projection of the head X Z PD and N are the categories of the complements The parentheses around Z mean Z is optional This doesn t work so well for S gt NP VP Q where is the head Linguistics 401 section 2 Hale October 15 2003 Phonology rules Q How are phonology and morphology related to one another the phonological form of words ACCOMODATES to the new environments created by morphology 1 Assuming that 941411 are allophones of a phoneme N l stem resent progressive stop kuro eruro press sand7 tayonrin ntayonrin spoil badga nibadga Yoruba spoken in Nigeria exhibits phonological place assimilation according to the new environment created by morphological pre xation 2 English plural su ixation allomorph phonological condition examples stem ends non sibilant voiceless C tap 7 taps ps tack 7 tacks stem ends in V or voiced non sibilant C tab 7 tabs bx sign 7 signs r11 tag 7 tags gz sigh 7 sighs following sibilant C kiss 7 kisses dish 7 dishes zz 7 zzes judge 7 judges 3311 sibilants a set of loud high pitched noisyfricatives77 p497 8 z j 3 1f 15 underlying form z morpheme occurs most widely Voicing assimilation voice a voice voice i Schwa epenthesis Q a 9 sibilant i sibilant bok z AnaetI z UNDERLYING FORMS 7 maetIez Schwa epenthesis bUkS Voicing assimilation boks mum SURFACE FORMS without this ordering r11ampgttf gts l 3 The Voicing Assimilation rule only applies in certain cases like not to face or case It is a morphophonological rule Morphology derivational rules add derivational a ixes etc in ectional rules add in ectional a ixes etc Morphophonology morphophonological rules apply to particular morphemes to determine which allomorph is used in each context Phonology phonological rules determine which allophones are used and make other changes in articulation that apply to all words Linguistics 401 section 2 Hale November 6 2003 Semantics binding and quantifiers 1 Binding theory Recall that we can consider a sentences I quot h on a particular interpre tation by specifying the intended coreference with subscripted indices Two classes of words exhibit contrasting referential possibilities possessive pronouns re exives my myself your yourself him himself her herself it itself our ourselves you yourselves them themselves 1 a Johni shaves Johni b gt1lt Johni shaves hirm c Himself shaves John d Johni shaves himselfi Names and nonreflexive pronouns are referential Reflexives are referentially dependent on another NP in the sentence Re exive binding is subject to standard agreement conditions gender animacy number 2 a gt1lt She shaved himselfi b She shaved herselfi c The furL shaved himselfi d The furL shaved itselfi e Wei shaved herselfi f Wt shaved ourselvesi gt an agreeing pronoun or name must precede the reflexive But the binder can t be too far away 3 a gt1lt Johni thinks 0p that Mary shaves himselfi b The barbersi wonder 0 who shaves themselvesi or too deeply embedded 4 a Johni worried himselfi b Np Pictures of Johm worried himselfi c Pictures of John b replicated themselves all over the computer screen c command Node A ccommands node B if an only if i A does not dominate B and B does not dominate 39 ii the first node i i A also i B Linguistics 401 section 2 Hale October 15 2003 Syntax more theta theory 1 Review 0 Just as grammatical categories D P V A are substitution classes of words so too are phrases substitution classes of word sequences 0 Phrases can occur in distinguishable positions before the verb complement of to or with special morphemes ga o indicating subjecthood in direct objecthood et cetera c There are general patterns to the assignment of theta 0 roles to argument phrases 11 Intransitive melt vs intransitive rule Some of these patterns are true of interesting subclasses of verbs Take melt for example 1 The sun melted the ice cream melted assigns THEME to object DP the ice cream 2 The ice cream melted melted assigns THEME to subject DP the ice cream The real world relationship between the scoop of Breyer s and the physical process by which solids become liquids is the same in 1 and 2 so we say the same theta role is assigned despite the ice cream77 occupying different syntactic positions This subclass of verbs the unaccusatives is vulnerable to a syntactic rule not unlike the dative shift rule we saw with passed Unaccusatives are a subclass of the verbs that have both transitive and intransitive forms 3 Iron Maiden rules the airwaves rules assigns AGENT to subject DP Iron Maiden 4 Iron Maiden rules rules still assigns AGENT to subject DP Iron Maiden intransitive melt assigns just THEME and intransitive rule assigns just AGENT 2 The 6 Criterion One effect theta roles have on the set of acceptable sentences in a language was famously articulated in Noam Chomsky s Lectures on Government and Binding7 1981 5 0 Criterion a Each obligatory 0 role selected by a predicate must be assigned to a referential expression eg DP b Each referential expression must be assigned a 0 role The unacceptability of 6 is then attributable to a violation of the 0 Criterion 6 Judy handed the mic What theta roles are assigned in the a examples and what s the problem with the corre sponding b s 7 7 a The dean s cronies implemented cutbacks b The dean s cronies implemented cutbacks to the chair 8 a Julie telephoned the donors b Telephoned the donors The Italian verb telephonare assigns the same theta roles and yet seems to be counterevi dence to the theta criterion 9 Ho telefonato has1sg telephoned l have telephoned 10 Gianni dice che ha telefonato G says that has3sg telephoned Gianni says that he has telephoned the morphology on the auxiliary verb suggests that it is agreeing with a pronoun that does not actually appear in the sentence 7 a zero morpheme PRO We can preserve the universality of Chomsky s 0 Criterion by postulating a Null Subject parameter which is set to YES eg Italian Spanish Chinese and set to NO eg English German French This is a parsimonious explanation in the sense that it can be re used elsewhere in the analysis of in nitival clauses 11 a W tried PRO to behave in a presidential manner b Karl told Dick PRO to behave himself c Colin is reluctant PRO to make a deal with North Korea d Hans abandoned the search for WMD PRO to avoid further hassle the explanation is that English in nitival clauses have the same parameter setting as ltal ian nite clauses Analogy even though German matrix clause are SV2OV German embedded clauses are SVO 7 just like English Linguistics 401 section 2 Hale October 15 2003 Syntax basic constituent order and argument structure 1 Word order phenomena There are an astronomical1 number of ways words can be sequenced Only some of them are recognizable as English What s the difference between the acceptable and the unacceptable sequences 1 review star means that native speakers say the sentence is Not part of their language grammatical category name of substitution class eg V A Adv N Name D P Prn Aux 2 Can you pick members of a grammatical category to make an acceptable sentence out of any of these randomly generated templates Name7Adv7A7Aux7Aux P7P7D7Aux7Prn Name7Prn7A7N7P A7Aux7Adv7Prn7N P7Name7P7A7V Aux7Adv7Prn7A7Aux P7Name7P7Adv7Name D7Name7A7P7Adv Adv7N7Aux7A7Aux 10 P7Aux7Aux7Prn7V PWNQWFPJN Which of these should have stars in front of them CNN will conduct three debates Bush has embarrassed some supporters 3 word substitutability Name 7 Aux 7 V 7 D 7 N Dennis might lose the 515551011 Carol should cast a vote Carol should cast a vote Name She should cast a vote Prn The woman should cast a vote determiner phrase DP The African American woman 4 phrase substitutability ShOUId caSt a VOte The one person who has a DP prayer of changing things in this country should cast a vote 1Using the nine categories from Fromkin there are 59 or about 19 million possible templates for ve word English sentences 5 DP is just the tip of the iceberg VP cast a vote rise to the occasion AP despondent about her boyfriend eager for victory NP candidate candidate for president PP about her boyfriend for victory CP that she s going to St Tropez if W will win Phrases XP headed by word of category X eg NPs don t have determiners at the beginning but rather nouns 2 Theta theory It would be nice for grammars to not only distinguish the acceptable and unacceptable sentences but also give some insight about how their structure contributes to their meaning To a rst approximation DPs refer to real or imagined entities in the world and verbs predicate properties of these entities Bush beats Clark AGENTincumbent president George W Bush 6 PATlENTformer General Wesley Clark Clark beats Bush AGENTformer General Wesley Clark PATlENTincumbent president George W Bush The same entities can participate in the same event in different ways We can describe this in terms of different thematic t9 roles AGENT PATlENT being assigned to different participants in the situation the sentence is about Shorthand 0 role T is assigned to argument phrase XP The speci c assignment depends on particular assigners and the particular environment 7 The recall thrills Schwarzenegger thrills assigns THEME to the DP The recall and EXPERIENCER to the Name Schwarzenegger 8 Schwarzenegger loves the recall loves assigns THEME to the DP the recall and EXPERIENCER to the Name Schwarzenegger 9 Judy passed the mic to Joe Lieberman passed assigns GOAL to the PP to Joe Lieberman and PATIENT to the DP the mic 10 Judy passed Joe Lieberman the mic passed assigns GOAL to the Name Joe Lieberman and PATIENT to the DP the mic English grammatical roles like subject direct object indirect object dative are distinguish able by position ln Latin Japanese Russian etc they are indicated by casemarking morphemes in a wide range of positions We have a budget version of this with the posses sive morpheme s 11 Schwarzenegger s handlers should have warned him about the media s assigns POSSESSOR to the Name Schwarzenegger Verbs are not the only 0 assigners predicative adjectives also do Linguistics 401 section 2 Hale November 4 2003 Semantics determiners presupposition 1 Determiners 1 No Beatles smoke I truth conditions No N VP is true just in case N 0 VP 3 I Beatles the set of Beatles B I VP smokes the set of smokers S Some other determiner truth conditions see others on p381 in Promkin 2 Every N VP is true just in case N C VP 3 Fewer than six N VP is true just in case VP lt 6 I in general the meaning of determiners is a relation between two sets N and VP I more specifically natural language determiners appear to be conservative relations conservativity A relation Q named by a determiner is conservative if and only if for any properties A and B relation Q holds between A and B if and only if relation Q holds between A and the things in A O B For conservative Q A QB is true just in case A Q AD B is true I We only have to look at the part of predicate B that overlaps with A in order to determine the truth of the whole sentence If human language determiner meanings are conservative then sentences whose truth conditions are AQB should be synonymous with sentences whose truth conditions are AQA O B constituent example kind of settheoretic object names WX Paul Wolfowitz individuals VPs snores occupies Iraq sets of individuals that do that activity Ns girl Beatle set of individuals that have that property As sleepy triggerhappy set of individuals that have that property NP that VP girl that sleeps Beatle that is happy set of individuals that have both properties A compositional semantic rule for a class of relative clauses 4 Np NP that VP NPE VP Some modifiers have this kind of intersective meaning But not all 5 a The girl who is wearing the blue cap bought a CD b The taller girl bought a CD 11 Are determiners really conservative doing the experiment What are the truth conditions of 6 6 Every Beatle smokes should be true just in case B C S is true in our model If every is conservative this sentence should be synonymous with a sentence having truth conditions B C B O S How can we make such a sentence 7 Every Beatle VP is a Beatle that smokes 8 Dpa Beatle that smokes B O S Indeed 7 has truth conditions B C B O S If every is really conservative then 7 should be synonymous with 6 Is fewer than 6quot conservative 9 Fewer than 6 Beatles smoke Nonconservative determiner meanings are de nable 10 owt is true just in case VI H gt 2 sentence truth value owt cars roll on the highway owt animals are muskrats owt bachelors are unmarried men owt teachers know modeltheoretic semantics owt teachers are teachers that know modeltheoretic semantics but not found in natural languages 2 Kinds of entailment Presupposition and Assertion assertions have meanings that are contradicted by their negation 11 a It is raining b It is not raining presuppositions are not similarly cancelled by negation 12 a I am thankful that it is raining b I am not thankful that it is raining Both sentences in 12 entail 11a 12a presupposes 11a What are the presuppositions of I regret that I lied to the queen 1 I regret that I lied to the queen 2 I lied to the queen 3 I am generally an honest person 4 There is a queen a A unique queen is identifiable in the context Linguistics 401 section 2 Hale December 3 2003 ls English a nite state language Formal languages In formal language theory a language is a set of strings A string is just a sequence of symbols chosen from an agreed upon set of symbols called the vocabulary or lexicon formal language non examples any sequence of English words from the OED E at by green maybe vanity all vanity 1 think que nous allons tromper the lesson strings over 1 b 0 starting with a abbb a a followed by a million b s bcacc the zero length string 6 mm strings over 1 b cd in alphabetical order 6 abd ad bed I abcd dbcd ha The idea of generative grammar is to use grammars to de ne a set that closely resembles a natural language 7 for instance all and only the acceptable English sentences However not all sets are de nable by all types of grammars What kind of rules do we require to accurately describe natural languages Finitestate grammars To de ne generate strings over 1 b c d in alphabetical order let S be the start symbol of the grammar in S gt a S1 S gt b S2 S gt c S3 S gt d S1 gt b S2 1 S1 gt c S3 S1 gt d S2 gt c S3 S2 gt d S3 gt d Grammar 1 has a corresponding nite state machine that recognizes all and only the sentences it generates The general form of nite state grammars Any grammar having only rules of the form A a 0 where A B are nonterminals and b is a terminal has a corresponding nite state machine Given a string if a path can be found through the machine the string is generated by the grammar and vice versa There are some languages that cannot be recognized by nite state machines formal language non examples sequence of a s followed by an equal number 6 ab aabb aaabbb of Us aabbb aaaaaaaaaaaabbb Call the number of a s in the sentence being analyzed 71 A nite state machine would need to remember this number 71 while waiting for the end of the Us But by de nition a nite state machine will only have enough states to remember some xed number of a s Hence there exists neither a nite state grammar nor a nite state machine for the language anb English is just like 1 Consider 2 a The cat died b The cat the dog chased died c The cat the dog the rat bit chased died d The cat the dog the rat the elephant admired bit chased died If A the cat the dogthe rat the elephant the kangaroo and B chased bit admired ate befriended it s clear that 2 have the structure 3 anbn ldied Chomsky and Miller 1963 argued that the obligatory paired dependencies presented by eitheror ifthen or the agreement between verbs and subjects can nest inside one another to an arbitrary depth 4 Anyonel who feels that if2 so manyg more4 studentS5 whom we6 haven ta actually admitted are5 sitting in on the course than4 ones we have thatg the room had to be changed theng probably auditors will have to be excluded isl likely to agree that the curriculum needs revision This is analogous to the reversal language mlem 6 ab which also cannot be gener ated by a nite automaton Grammar 5 does generate the language anb It is a contextfree grammar and as such is capable of deriving any number of center embeddings S gt a b 5 S a a S b There is a corresponding machine that accepts the context free languages the push down automaton Linguistics 401 section 2 Hale November 1811 2003 Human sentence processing ambiguity 1 Global ambiguity S NP VP NP gt the spy NP gt the cop NP gt NP PP Rayner Carlson and Frazier 1983 PP gt with binoculars VP gt saw NP VP gt mm NP PP Figure 1 PP attachment grammar modifiers are sisters I noun phrase adjunction ruler mo ifies this NP verb phrase likewise has optional prepositional phrase PP complement Syntactic analyses suggest semantic content 1 S NP the spy VP saw NP the cop PP with binoculars binoculars help the spy do his seeing 2 S NP the spy VP saw NP NP the cop PP with binoculars area quot L ofthecop Consider the case of a PP modifying an NP 3 3 Put the block in the box on the table 3 Put the block in the box on the table 4 so Put the block in the box on the table in the kitchen 3 Put the block in the box on the table in the kitchen C Put the block in the box on the table in the kitchen 2 Put the block in the box on the table in the kitchen 3 Put the block in the box on the table in the kitchen As the number of PPs increases the number of possible analyses increases the Catalan numbers Cat 25144213246914304862 2n 2n n n 1 Church and Patil 1982 2 Local ambiguity We take in words serially one at a time and yet we have some idea what people mean before they finish speaking For the human sentence processing mechanism to work in this eager fashion assumptions need to be made about the intended structure before the end of the sentence 5 The dog walked to the park had been chewing the bone If these structural assumptions turn out to be wrong the processor has been led up the gar den path Sentence like 5 are known garden path sentences because of their temporary ambiguity 6 TP DP TI D N TVp V PP walk P DP to the park 7 TP DP T T VP DP JP l had D N 1 l l The dog 1 TP l empty walked to the park H the band started in suburban Chicago gained popularity in New York 5 the nurses switched to overnight duty became irritable during the day 9393 the inmates crowded in the hallway had an air of misery 5 the vendors closed for winter will reopen in the spring 3 Methods A Garden Path effect is the difference between the temporarily ambiguous sentence and an unambiguous control sentence 8 a The dog walked to the park had been chewing the bone b The dog that was walked to the park had been chewing the bone self paced reading subjects take much more time at had been chewing eye tracking subjects look longer in the region had been chewing look back to earlier regions the same methods can be used to investigate sentences that are grammatical yet abnormally di iculty to understand the reporter disliked the editor 9 539 the reporter who the senator attacked I disliked the editor 1 the reporter who the senator who John met I attacked I disliked the editor 10 V John met the senator who attacked the reporter who disliked the editor I I 4 The Garden Path Model of human sentence processing Lyn Frazier and Janet Podor s 1978 Garden Path Model proposed two heuristic prin ciples that seem to guide the human processor s choice of phrase structure rule Minimal Attachment Do not postulate any potentially unnecessary nodes Late Closure aka Right Association If grammatically permissible attach new items into the clause or phrase currently being processed Consequences of Minimal Attachment I The main verb reading in 6 requires extra TP DP CP and nodes compared to the reduced relative The feeling of confusion at had been chewing is because the parser must go back and reanalyze at that point I The PP attachment ambiguity from grammar 1 is explained by that fact that the VP attachment ie VP gt saw NP PP uses one less layer of NP recursion and is therefore preferred Consequences of Late Closure I The direct object reading vs intransitive jogsquotquot is correctly predicted for 11 Since Jay always jogs a mile this seems like a short distance to him versus 12 Since Jay always jogs a mile seems like a short distance to him I Low attachment is correctly predicted for adverbs like yesterday 13 Joyce said Tom left yesterday versus 14 Joyce said E M CZ yesterday Linguistics 401 section 3 Sentence processing October 25 2007 1 Sentence Processing Syntax tells us that sentences have a structure but it doesn39t tell us how that structure is used Is this one of syntax39s downfalls No Syntax is concerned with linguistic knowledge and does not postulate what constitutes the behavior of individuals when using this knowledge Sentence processing however is concerned with language users39 behavior The parser or human sentence processor uses the grammar as well as additional rules to guide the comprehension of sentences It has the following properties incremental Builds structure as the sentence is being heard It does not wait until the full sentence has been heard to begin building structure lefttoright Builds structure from left to right robust Accurately processes previously unseen data sensitive to prior It can be trained linguistic experience informed by grammar The parser can not be fully explained by means of statistics or heuristics but requires a component that specifies the grammar of the language Does syntactic structure affect sentence processing Does sentence processing affect syntactic structure 2 Ambiguity Sentences can have more than one syntactic structure as shown by the following real newspaper headlines a HERSHEY BARS PROTEST b CHOU REMAINS CREMATED c ENRAGED cow INJURES FARMER WITH AX 1 Structural ambiguity leads to different meanings for the same string of words as shown by the prepositional phrase PPattachment ambiguity in 2 for the sentence quotJohn bought the book for Susanquot N39 vlNP PP R F Det N PP Det bought the book for Susan bought the book for Susan 2 In 2a the meaning of the sentence would be interpreted as quotJohn bought the specific book for Susanquot nominal complement whereas in 2b the meaning would be quotJohn bought a book and the book was for Susanquot verbal adjunct The meaning difference is reflected by the different syntactic structures This leads to complications If there are multiple possible interpretations for a sentence how do we select the correct one Is there a preference for one reading over another 3 Garden pathing Some sentences have a relatively simple syntactic structure but they39re very difficult for humans to understand as in 3 a The doctor sent for the patient arrived b The old train the young c The cotton clothing is made of grows in Mississippi 3 Frazier39s 1979 garden pathing theory specifies that this difficulty results from the parser building an incorrect analysis that it can39t recover from For instance when interpreting 3a the parser wants to make quotthe doctorquot the subject of the verb phrase quotsent for the patientquot as in 4a But when it gets to the word quotarrivedquot it crashes because it can not attach this word anywhere 4b shows the correct analysis in the sentence which corresponds to a reduced form of the phrase quotThe doctor who sent for the patient arrivedquot a Mainverb reading b Reducedrelative reading IP IP I 39 NP I P I VP NP IP I VP PSt Pst the doctor sent for the patient arrived the doctor sent for the patient arrived 4 In other words 4a shows the syntactic structure the parser is building as it is being led down the garden path of the incorrect analysis Syntax can not explain why these sentences are unacceptable for most speakers ie the structure in 4b is perfectly grammatical This is a performance issue and should be explained by the rules and procedures of the parser itself Some garden paths are more temporary than others For example native English speakers would not characterize the sentence in 5 as ungrammatical John knew the answer was wrong 5 In 5 the NP quotthe answerquot is the subject of the verb quotwasquot However eyetracking studies indicate there is an increase in processing time when the word quotwasquot is encountered Ferreira amp Henderson 1990 signifying that the parser has some difficulty 6 shows the average total reading time results for the different regions of subjectobject ambiguity sentences based on Ferreira amp Henderson39s study John knew the answer was wrong 26 276 ambiguous region disambiguating region 6 This increase in processing time would be further supported by the hypothesized incremental interpretation of the sentence quotJohn knew the answer was wrongquot shown in 7 a b IP C IP NP NP 39 NP 39 A A A VP VP John John pst I John pst I V39 V39 I v V I NP l A d W knew the answer NP l39 A VP John PSt I V V IP knew the answer was 7 According to Frazier 1979 this data can be explained by the following heuristics attributed to the human sentence processor minimal attachment Don39t postulate new nodes unless you have to eg 4b late closure Attach new words to the clause currently being processed eg 3b canonical sentoid strategy Interpret NVN as SVO eg 7c Fodor et al 1974 By employing heuristics like these the parser is generally able to quickly and easily build correct structures for the majority of sentences Garden path sentences fall outside this domain of acceptability


Buy Material

Are you sure you want to buy this material for

25 Karma

Buy Material

BOOM! Enjoy Your Free Notes!

We've added these Notes to your profile, click here to view them now.


You're already Subscribed!

Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'

Why people love StudySoup

Steve Martinelli UC Los Angeles

"There's no way I would have passed my Organic Chemistry class this semester without the notes and study guides I got from StudySoup."

Jennifer McGill UCSF Med School

"Selling my MCAT study guides and notes has been a great source of side revenue while I'm in school. Some months I'm making over $500! Plus, it makes me happy knowing that I'm helping future med students with their MCAT."

Jim McGreen Ohio University

"Knowing I can count on the Elite Notetaker in my class allows me to focus on what the professor is saying instead of just scribbling notes the whole time and falling behind."

Parker Thompson 500 Startups

"It's a great way for students to improve their educational experience and it seemed like a product that everybody wants, so all the people participating are winning."

Become an Elite Notetaker and start selling your notes online!

Refund Policy


All subscriptions to StudySoup are paid in full at the time of subscribing. To change your credit card information or to cancel your subscription, go to "Edit Settings". All credit card information will be available there. If you should decide to cancel your subscription, it will continue to be valid until the next payment period, as all payments for the current period were made in advance. For special circumstances, please email


StudySoup has more than 1 million course-specific study resources to help students study smarter. If you’re having trouble finding what you’re looking for, our customer support team can help you find what you need! Feel free to contact them here:

Recurring Subscriptions: If you have canceled your recurring subscription on the day of renewal and have not downloaded any documents, you may request a refund by submitting an email to

Satisfaction Guarantee: If you’re not satisfied with your subscription, you can contact us for further help. Contact must be made within 3 business days of your subscription purchase and your refund request will be subject for review.

Please Note: Refunds can never be provided more than 30 days after the initial purchase date regardless of your activity on the site.