LITERARY THEORY ITALIAN 0270
Popular in Course
Popular in Italian
This 234 page Class Notes was uploaded by Mrs. Roberta Christiansen on Friday September 4, 2015. The Class Notes belongs to ITALIAN 0270 at University of California - Los Angeles taught by Staff in Fall. Since its upload, it has received 124 views. For similar materials see /class/177328/italian-0270-university-of-california-los-angeles in Italian at University of California - Los Angeles.
Reviews for LITERARY THEORY
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: 09/04/15
SESSION 3 COMINGOUT TO MYSELF OBJECTIVES 1 Participants will understand that the phases they are experiencing in coming to terms with their gay and lesbian orientations are typical and need not be feared 2 Participants will improve their ability to identify the harm that can come to their views of them selves from denying their gay and lesbian orientations 3 Participants will improve their skills in resolving confusion over inaccurate information feelings adult reactions rejection and both myths and stereotypes 4 Participants will identify for themselves culturally defined gender roles 5 Participants will identify and present their reasons for being proud to be gay lesbian or bisexual RATIONALE Adolescence is a time of experimentation when developmental tasks are faced One such task relates to trying out social roles Another task involves learning social skills Others deal with sexuality and identity Coming to terms with sexual orientation through discovery and acceptance is among the more important objectives to be achieved during this period Because emerging feelings toward the samesex are often startling and troublesom eggoing against conventional normsigay and lesbian youth experience additional kinds of stress and confusion than that experienced by straight youngsters The stigma associated with homosexuality prevents clarifying and resolving many critical issues First the methods for problem solving available to other teenagers are not there for gay and lesbian teenagers Models are miss39 While heterosexual youths talk to parents other adults and peers open exploration of sexual orientation questions with family and friends is simply not feasible for gay and lesbian adolescents Second not being able to talk to parents other adults or peers and having difficulty admitting their orientation to themselves these youths become isolated This isolation plus the stigmatized attitudes and pictures of gays and lesbians which are available often leads to relying on stereotyping and misinformation Believing the negative images of gays and lesbians results in low selfesteem shame and selfhatred This training session is designed to provide a limited opportunity to work on the confusion surrounding gay and lesbian identity in adolescence Next the session is designed to enable these young people to feel more comfortable accepting their sexual orientation and to develop a sense of pride Without working through these issues involved with sexual orientation instruction in safer sex is difficult and often perceived as tangential to their concerns and priorities PROCEDURES 1 Introduction of group members and the topic of the day 2 Check Feeling Thermometers 3 Review of successes that occurred between sessions 4 Giving out tokens 5 Use of a script and discussion to recognize the natural steps of denial rejection and depression when initially confronted with recognizing one is gay or lesbian 6 Use of two role plays to explore the damage one can to do oneself from denying one39s sexual orientation and share what happened when recognition of one39s orientation rst took place 7 Reduce confusion over misinformation false expectations and stereotypes through practice cards which place the participant in the position of correcting mistaken views 8 Discover culturally defined gender role characteristics through small group work and role play issues 9 Gain pleasure in selfacceptance as each participant goes through the ritual of standing before the group and indicating why heshe is proud of hisher identity and then receiving a certificate MATERIALS Tokens Feeling Thermometers Scripts of Jack and Lefty Maria and Grace Instruction Card Jenny and Gail Instruction Card Ted and Abe Practice cards on Confusion Form Masculine and Feminine Behavior Pencils Blank Cards Certificates Exercise 1 Introducing Each Other and the Session GROUP LEADER SAYS Welcome back Today39s session is about comingout to ourselves We are going to practice acknowledging our sexual orientation and handling how we feel about knowing we are gay or lesbian Before we go any further I want to know how you would complete this sentence quotI like being gay or lesbian because Please get into a group for the men and one for the women DIVIDE THE GROUP INTO A GAY GROUP AND A LESBIAN GROUP Take three minutes and brain storm a list of answers to that question quotI like being gay or lesbian because quot For example you might say quotbecause I have so many caring friends quot ALLOW THREE MINUTES AND ASK FOR THEIR LISTS What did you come up with OBTAIN IDEAS OTHER LISTS HAVE CONTAINED THE FOLLOWING BECAUSE MY FRIENDS ARE CREATIVE GAY AND LESBIAN PEOPLE ARE NICE TO ME I AM A TRUE ORIGINAL I CAN BE EXPRESSIVE IT MAKES ME STRONGER I CAN BE TRUE TO MYSELF KNOW HOW TO BE OPENMINDED MY FRIENDS ARE MORE SENSITIVE I JUST FEEL FREER WE GAYS AND LESBIANS ARE LEADERS WE STICK TOGETHER AND SUPPORT EACH OTHER That39s a wonderful list Feeling good about being gay or lesbian is really important Speaking of quotfeelingquot how do you feel now on the Feeling Thermometer OBTAlN READINGS First I want to make sure everyone knows who is in the group As we go around please tell us your first name and tells us about a time you did something really brave It could be many things overcoming a fear speaking out saying quotnoquot to drugs confronting someone telling someone a personal secret opening up to another person leaving a bad scene admitting you were wrong trying something again and again helping someone when other people looked down on you for doing it meeting a new person and so on To give you an idea of what I mean I39ll go first I39m your group leader and my name is I did something brave whenI GIVE A REAL EXAMPLE HAVE EACH PERSON GIVE HISHER NAME AND A TIME WHEN SHEHE DlD SOMETHING BRAVE IF SOMEONE HAS TROUBLE THINKING OF AN INCIDENT HELP THEM DISCOVER A TIME OF BRAVERY LIKE JUST COMING TO THIS SESSION MAY BE A BRAVE ACT FOR A TEENAGER HAVE THE COGROUP LEADER GO LAST Here are some tokens for you to pass out to other group members during this session Every time you appreciate any act of another person or feel positive toward them or even simply like their being here give them a token PASS OUT TOKENS TO GROUP MEMBERS Now let39s take a few minutes and tell us what you did between sessions that kept you away from the HIV virus or that made your life a little bit better GO AROUND AND HEAR REPORTS GIVE OUT TOKENS Those reports were really great What about the homework You were to try concentrating on a simple taskilike washing your face brushing your teeth Could you really focus on the taskgget into it so much that your forgot yourself at that moment ENCOURAGE RESPONSES AND DISCUSS As I said earlier today we39ll deal with comingout to oneself As you know the goal of all these sessions is to help you reduce your exposure to HIV If you already are HIV positive then the goal is to help you live a high quality life and for you to learn how to keep from spreading the virus around But we don39t think it is very easy to learn about HIVAIDS without first exploring what it means to be a gay or lesbian teenager In this session we39ll work on comingout to yourself and in the next session coming to other people will be the topic Exercise 2 What39s Wrong With Me When a person is going through a deep struggle within him self or herself it is easy to feel all alone It is also easy to think that quotthere must be something wrong with me It is easy to think you are really different Those thoughts and feelings are magnified for the young person recognizing that he is gay or that she is a lesbian Society39s lack of acceptance makes things more confusing This little script will help illustrate whatI mean Who will read Jack and who will read Lefty SELECT VOLUNTEERS AND GIVE THEM THE SCRIPTS Jack and Lefty LEFTY I want to ask you something JACK What39s that LEFTY Are you queer JACK Man You think I39m crazy LEFTY No buggin Just answer me Are you JACK LEFTY JACK LEFTY JACK LEFTY JACK LEFTY JACK LEFTY JACK LEFTY JACK LEFTY JACK LEFTY JACK LEFTY JACK Hell no Why you ask I don39t see you going out You never try to put a move on a girl You don39t talk about sex There39s no pictures of women in your room Give me a break I39m no faggot I hate faggots If some queer came up to me I39d beat him up You really hate them You better believe it You39d beat them up Damn right You39d beat me up You39re no queer You like me What is this Just answer meido you like me Course I doinot like a queer though I know you Jack Don39t hide from me I39ve been watching you for a year Jack When you going to admit it I39m gay When you going to say you39re gay No no no Don39t be all upset Be happy you can finally talk to somebody I wish I were dead Man don39t be stupid I don39t want to be gay THE END Jack and Lefty are casual friends They met in high school and have known each other for several years Before we begin reading the script I want to give each person who is observing something specific to look for ASSIGN EACH PARTICIPANT A DIFFERENT TASK You observe their facial expressions v01ce tone posture words gestures voice level eye contact OK now let39s read the script HAVE THE VOLUNTEERS READ THE SCRIPT Thank you That was great GIVE OUT TOKENS TO THE ACTORS Let39s hear what people observed OBTAIN RESPONSES LOOK FOR DENIAL TURNING AGAINST HOMOSEXUALITY AND FEELING VULNERABLE IN JACK VULNERABILITY IS SEEN lN FEELING BAD ABOUT ONESELF AND FEELING EASILY HURT LOOK FOR SUPPORT AND CONCERN FROM LEFTY GIVE OUT TOKENS So Jack was obviously distressed about having Lefty suggest that he was gay and what did he do to deal with it ENCOURAGE RESPONSES That39s right first he denied it Then he rejects gayness and takes on some of the hatred of the dominant society Feelings of depression and wanting to harm him self come out at the end Was Jack crazy or unusual in responding as he did ENCOURAGE DISCUSSION While there are many gay males and lesbian women who feel good about themselves most have felt distressed at some point during the com ingout process Lesbian and gay people are gaining their rights but society is far from being tolerant and accepting With the world around us considering being gay or lesbian a crime a sin or sign of mental illness it is not surprising that someone would go through the stages we saw here idenial of feelings irejecting people who are gay or lesbian and ifeeling depressed and suicidal The point is that these are normal responsesinot something off the wall Society has the problem not gay and lesbian youth There is nothing wrong with having these reactions during the selfdiscovery part of comingout How are you feeling now on the Feeling Thermometer OBTAIN FEELING THERMOMETER READINGS Exercise 3 quotI Don39t Want To Be Gav 01 Lesbian quot During the struggle of comingout some gay and lesbian youth want to forget who they are or want to be heterosexual Let39s see how we would deal with that situation through a role play Who will be Maria and who will be Grace SELECT VOLUNTEERS You two are friends Grace wants to forgetito erase from her mindithat she is a lesbian Grace what is your goal HAVE GRACE STATE WHAT HER GOAL IS AND HELP HER BE CLEAR IF SHE IS OFF Maria your goal is to convince Grace that quotforgettingquot it might not be very helpful in the long run So Maria what is your goal HAVE MARIA STATE HER GOALiTO CONVINCE GRACE THAT TRYING TO CHANGE WHO SHE IS HAS NEGATIVE CONSEQUENCES AND HELP HER BECOME CLEAR IF SHE IS OFF TARGET Maria I am going to give you a card with a few ideas on it GIVE MARIA THE quotINSTRUCTION CARD quot INSTRUCTIONS FOR MARIA YOUR TASK IS TO CONVINCE GRACE THAT TRYING TO FORGET WHO SHE IS CAN END UP DOING HER NO GOOD SOME POSSIBLE lDEAS ARE AS FOLLOWS l CUTTING OFF PART OF YOUR SELF WILL LEAVE YOU WITH A DAMAGED SELFgONLY PART OF YOU WILL EMST 2 GIVEN THAT YOU CAN NOT CHANGE WHO YOU ARE YOU WILL END UP FRUSTRATED BY TRYING TO MAKE THE CHANGE 3 YOU ARE GONG TO END UP CONVINCING YOURSELF THAT THERE IS SOMETHlNG WRONG OR HATEFUL ABOUT BEING A LESBIAN ADD YOUR OWN IDEAS USE ANY IDEAS THAT YOU WANT Maria feel free to use your own ideas You don39t have to follow what is on the card The rest of you will be observers Each person gets a specific task GIVE EACH PERSON SOMETHING TO FOCUS ON You observe their facial expressions voice tone posture words gestures voice level eye contact OK let39s get started with the role play ALLOW THE ROLE PLAY TO GO FOR A FEW MINUTES Thanks That was really good GIVE OUT TOKENS Grace where were you on the Feeling Thermometer during the role play OBTAIN A RESPONSE Also Grace what did you do that you liked and what would you do differently OBTAIN A RESPONSE Maria where were you on the Feeling Thermometer during the role play OBTAIN A RESPONSE Maria what did you do that you liked and what one thing would you do differently OBTAIN A RESPONSE Now let39s hear from the observers Tell us what you observed what you liked and what one thing you would have done differently if you had been playing the role OBTAIN FEEDBACK FROM THE OBSERVERS LOOK FOR INCREASING SELFHATRED ENDlNG UP IN THE WRONG CROWD OF PEOPLE STLL HAVING TO DEAL WITH ONE39S TRUE FEELINGS AND BEING SO INVOLVED WITH DENIAL THAT A LOT OF FUN IN LIFE IS MISSED ALSO GIVE OUT TOKENS Are there any other arguments that you would have used if you were Maria OBTAIN RESPONSES Now let39s do one more role play Who will be Jenny and who will play Gail SELECT VOLUNTEERS Jenny you are a lesbian and your task is to explain to Gail Why it is best for you to act as if you were heterosexual Tell us What your goal is Jenny HAVE JENNY STATE THAT HER GOAL IS TO CONVINCE GAIL THAT ACTING AS IF SHE WERE STRAIGHT IS THE BEST THING FOR HER JENNY TO DO Gail your task is to point out to Jenny Why pretending she is heterosexual might not be a good idea What is your goal Gail HAVE GAIL EXPLAIN THAT HER GOAL IS TO TELL JENNY WHY ACTING LIKE A HETEROSEXUAL IS NOT A GOOD IDEA Gail here are some ideas that you might want to use GIVE OUT AN quotINSTRUCTION CARD TO GAIL INSTRUCTIONS FOR GAIL YOUR TASK IS TO CONVINCE JENNY THAT HER PLAN TO ACT LIKE SHE IS HETEROSEXUAL MIGHT HAVE A NEGATIVE IMPACT ON HER YOU MAY WANT TO 1 ASK HER IF SHE THINKS IT IS POSSIBLE TO BE A LESBIAN AND STILL THINK OF HERSELF AS A WOMAN AND MEET MANY OF THE EXPECTATIONS THAT SOCIETY HAS OF WOMEN 2 HELP HER THINK THROUGH WHAT CHANGES SHE E 3ECTS IN HER LIFESTYLE FROM quotACTING HETEROSEXUAL AND SEE IF THEY ARE REALLY GOlNG TO HAPPEN 3 HELP HER SEE WHO IS PUTTING PRESSURE ON HER AND THINK OF WAYS TO DEAL WITH THEM 4 HELP HER THINK THROUGH WHAT SHE E 3ECTS FROM BEING A LESBIAN AND SEE IF THERE ANY MYTHS PRESENT 5 MAKE SURE THAT SHE IS NOT BLAMJNG BEING LESBIAN FOR ALL THE UNPLEASANT THINGS IN HER LIFE USE ANY IDEAS THAT YOU WANT MAKE SURE YOUR IDEAS ARE INCLUDED Gail feel free to use your own ideas You don39t have to follow What is on the card The rest of you will be observers Each person gets a specific task GIVE EACH PERSON SOMETHING TO FOCUS ON You observe their facial expressions voice tone posture words gestures voice level eye contact OK let39s get started with the role play ALLOW THE ROLE PLAY TO GO FOR A FEW MINUTES Thanks That was really good GIVE OUT TOKENS Jenny where were you on the Feeling Thermometer during the role play OBTAIN A RESPONSE Also Jenny what did you do that you liked and what one thing would you do differently OBTAIN A RESPONSE Gail where were you on the Feeling Thermometer during the role play OBTAIN A RESPONSE Gail what did you do that you liked and what one aspect would you do differently OBTAIN A RESPONSE Now let39s hear from the observers Tell us what you observed what you liked and one thing you would have done differently if you had been playing the role OBTAIN FEEDBACK FROM THE OBSERVERS LOOK FOR ASSUMING ALL PROBLEMS ARE SOLVED IF ONE IS HETEROSEXUAL AND DEALING WITH STEREOTYPES OF LESBIANS ENCOURAGE A DISCUSSION OF GENDER ROLES HOW ARE MEN AS MEN SUPPOSED TO BEHAVE AND HOW ARE WOMEN AS WOMEN SUPPOSED TO BEHAVE7 ALSO GIVE OUT TOKENS Are there any other arguments that you would have used if you were Maria OBTAIN RESPONSES Think of the day on which you came to the conclusion that you were gay lesbian or bisexual Where were you What did you have on Who were you with How old were you What time of day was it What did you say to yourself What did you do Think a few minutes about that day ALLOW A FEW MINUTES FOR REMEMBERING THEN DIVDE THE GROUP lNTO PAIRS Now describe that day to your partner ALLOW A FEW MINUTES FOR SHARING How do you feel on the Feeling Thermometer OBTAJN READINGS How did the remembering and sharing go OBTAIN REACTIONS AND DISCUSS WHAT WAS POSITIVE AND NEGATIVE ABOUT COMINGOUT TO YOURSELF Exercise 4 Clearing Up Confusion Often times comingout to oneself is made harder by confusion over the facts of gay and lesbian development ways of behaving feelings and sexual experiences To get at some of the confusion and clear it up we are going to use the practice cards You will receive a card and at the top it tells you to pass the card to a certain person Pass the card and the person who gets it will read you a statement Then you indicate whether you agree or disagree with the statement and tell us m We will then ask the other group members for their opinions PASS OUT THE CARDS ONE AT A TIME AFTER A RESPONSE HAS BEEN MADE TO A CARD ASK THE GROUP39S OPINION AND DISCUSS IT FOCUS ON MISINFORMATION AND FACTORS THAT CONTRIBUTE TO CONFUSION AROUND EMERGlNG HOMOSEXUAL FEELINGS SUCH FACTORS MAY INCLUDE MYTHS AND STEREOTYPES INABILITY TO LABEL FEELINGS ADULTS DISMISSING FEELINGS CONFLICTS ABOUT HOMOSEXUALITY ASSUMPTIONS REGARDING THE ORIGINS OF HOMOSEXUALITY AND NORMATIVE BEHAVIORS CONFUSION CARDS CARD 1 S CONFUSION PASS TO THE PERSON WHO LIKES CHINESE FOOD COOamp EISELNVEIEIT SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S 9 CIHVC JAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO EIEIHDV IIOA OC 39EISVHcI SIHI JO HVEIA ISVT EEHI EIEI CFHIOHS HVEIA IXHN 39AHHOM OI ION 39MON LI ILI 39II JO IIIO MOHD CFHIOM I 39EISVHcI V lelf SVM II CIVS EIHS NO 3W GELNHIII SAOEI HEEHIO IVHI IIOII AW CquotIOI I VI SVM I NEIHMH COOamp IIIOS SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S S CIHVC JAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO EIEIHDV IIOA OC LI NVoiNVIEISEIT V EIEI INVC I OS HAISSEIHDDV IOV HO NVII V EDIH SSEHC OI ELIVH I IIIEI 39STHID HEIHIO OI GEIIOVHIIV ATTVIIXEIS quotIEIEM I CLNV quotIHID CquotIO HVEIA NEIEIIiII ELNINIWEM CLNV AIIHHCI AHEIA V ILIH COOamp HEIEIHD SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S 17 CIHVC JAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO HEIHDV IIOA OC OC SIHDIVHIS SV SDNITEIEM EEWVS EIHI EIAVH LNVC SAVD SSEHID I 39CIHIHM SCLNIIOS EIAOT NI DNITIVJH ILI AVS OI IIIEI 39CIHVMCEI EDIH ATTVEIH I CLNV EICLIIVTO SI EEWVN AW COOamp NVEIHOH SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S E CIHVC JAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO HEIHDV IIOA OC 39EEW HIIM DNOHM DNIHIEWOS EIEI ISHW EIHEEHI 39CESIIOHV ATIVIIXEIS EIHHM AEIHI NEEHM IIIOEIV gtIquotIVI STHID IHDIVHIS IVHM EDIH DNIHIANV CUIOS LNOG AEEHI 39THID HEIHIONV HO SDNITEIEM quotIVIIXEIS AW EIEIIHOSEIG LNVC In COOamp NVITVII SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S 3 CIHVC JAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO HEIHDV IIOA OC 39AVD EIEI INVC I OS 39ION ILI 39AHSIMS EIHV SAVD IIIEI 39SAIID HEIHIO OI GHIOVHIIV quotIEIEM I CLNV ELLEITHIV OHOVII AHEIA V ILIH COOamp EISHNVCIVI SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOO S 11 CIHVC AAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO EIEIHDV IIOA OC 39HVEIA IXHN NIVDV II AHI quotIII IIIEI HEM EDIIT LNCIC I 39EIIIIIIISOHCI V OI OD OI EW CquotIOI EEH 39WEIHI HOIIS LNOC lelf AEIHI 39SEISINEICI SNEEW HEIHIO IIIOEIV HNIHI NEW TIV CIVS EIH 39AVD SVM I IHDIIOHI I CVC AW CquotIOI In COOamp NVOIXHW SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOO S 01 CIHVC AAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO HEIHDV IIOA OC 39NVIEISEIT V EEWOOEIEI OI IIIEI EDIOHC ON CIVH I IVHI HEIchV 39MONgtI LNCIC I EW IIIOEIV DNIHIEWOS GEISNEIS EIAVH lelW EIHS 39II EIAEIITEIEI LNVC I 39GEISIIOHV OS 109 I 39NEIAEFIEI SVM I NEIHM EW GHOIIGEIS NV IEISEIT HEICFIO SIHIH SEHIOEIEIHVEI SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S 6 CIHVC AAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO HEIHDV IIOA OC n 39AVD WAI AHM S lVHI 39HIIIOW AW N SINEIcI SIH EDIVI EW EICIVW AquotIIWVamp EIHI JO CLNEIIHJ SIHI CIgtI V SVM I NEEHMH COOamp NOIVC SEDIH OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S 8 CIHVC LAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO EIEIHDV IIOA OC AI39GEIHIVH DNIHEI TIIM EIAOT IVHI HNIHI OI DNISIIiNOO S II 39OOI EW OI quotIHIIHO EIEI CLAEIHI HHH MONgtI OI 109 I HI IIIEI DNIIIOXEI SAEIHS HNIHI I 39HEIH OI NVEEW EIHV SCIgtI 39EHZZEIT V SquotIquotIVC HNO AHHAEI WOHM EIGVHD AW NI THIS V SIEIHEEHIH COOamp EIEDINVA SEDIIT OHM NOSHEIcI EIHI OI SSVcI NOISIIJNOC S L CIHVC AAHM CLNV DNDINIHI SI NOSHEIcI SIHI AVM EIHI HIIM HEIHDVSIG HO HEIHDV IIOA OC 39quotITV HELIHV HOIIIAI IVHI WIH INVM LNVC I SSELIID I 39SHEEHIO quotHELL CLNV EW IOEHEIH TIIM EEH WIH HOVOHcIcIV I HI 39SSVquotIC HSITDNEI AW NI AOEI SIHI INVM ATIVEIH I CLNV NIAEDI SI EEWVN AW quotI39VE HAD SEX WITH ABOUT TWENTYFIVE GIRLS IN THE LAST THREE MONTHS THERE AREN39T TOO MANY MORE GIRLS LEFT IN MY HIGH SCHOOL WHO WILL DO IT I GUESS THAT SHOULD FIX ANYONE WHO THINKS THAT I39M GAY DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY CARD 12 S CONFUSION PASS TO THE PERSON WHO LIKES INDIAN FOOD quotI CAN HAVE AN ORGASM WITH A GUY OR ANOTHER GIRL I LIKE SEX THAT39S ALL IF THEY HAVE THE RIGHT MOVES WHO IS HAVING SEX WITH ME DOESN39T MAKE A BIG DIFFERENCE SO THAT lIUST MAKE ME SOME KIND OF FREAKquot DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY Quickly tell me how your Feeling Thermometer levels would read right now OBTAIN READINGS AND IN WHAT WAY THE READINGS RELATE TO THE CARDS AND SITUATIONS So if you had to summarize it what is a positive message to give yourself when this kind of confusion comes up ENCOURAGE RESPONSES LOOK FOR HELPFUL AND POSITIVE SELFTALK Thank you for your responses and contributions Are there areas of confusion that we have not touched upon ENCOURAGE A RESPONSE AND DISCUSS Exercise 5 Identifving Gender Roles Now I want you to consider how men and women typically behave In your society what is considered masculine and what is considered feminine We know that these definitions change depending on the culture you come from Let39s go over the form I want you to ll out in small groups HAND OUT THE quotMASCULINE AND FEMININE BEHAVIORquot FORM AND GO OVER IT MASCULINE AND FEMININE BEHAVIOR 1 LIST THE WORDS THAT DESCRIBE TYPICAL MASCULINE AND FEMININE BEHAVIOR THAT IS ACCEPTED IN YOUR COMMUNITY MASCULINE BUTCH FEMININE FEM 2 LIST THE WORDS THAT DESCRIBE HOW MEN AND WOMEN BEHAVE IN DEEP INTIMATE RELATIONSHIPS MEN WOMEN 3 LIST THE WORDS THAT DESCRIBE HOW MEN AND WOMEN BEHAVE SEXUALLY MEN WOMEN THE END BREAK THE LARGE GROUP INTO TEAMS OF THREE PEOPLE IF IT IS POSSIBLE CREATE CULTURALLY HOMOGENEOUS GROUPINGiBLACKS TOGETHER LATINOS TOGETHER ETC Go ahead and take about 10 minutes to fill this out ALLOW TEN MINUTES AND THEN HAVE THE GROUPS REPORT ON THEIR LISTS LOOK FOR CULTURAL DIFFERENCES DISCUSS THESE DIFFERENCES AS THEY AFFECT HOW A GAY OR LESBIAN PERSON FEELS IN THAT ENVIRONMENT That was great I want two volunteers to read this scene The script starts with giving you lines but then you have to keep going on your own making up the lines Who will play Ted and who will play Abe SELECT VOLUNTEERS AND GIVE OUT THE SCRIPTS Ted and Abe TED This will be our first weekend together I39m so excited ABE Me too I got tickets to a play for tonight TED Really I got us tickets for a new dance group for tonight ABE I39m used to being the one who gets tickets TED Me too Well I39ll cook I39m a good chef ABE Look usually I39m the chef You do the dishes TED Men don39t do dishes ABE When we get in bed it won39t matter TED I39ll let you know when I39m ready ABE I39m used to being the one who starts it TED Well at least this is definite I39m a top ABE You can39t be because I have always been the top TED KEEP GOING TRY TO RESOLVE THIS ON YOUR OWN ABE KEEP GOlNG TRY TO RESOLVE THIS ON YOUR OWN THE END While you two read the parts the rest of the group will be observers When the role play is over you can give some feedback ASSIGN OB SERVER ROLES You observe their facial expressions voice tone posture words gestures voice level eye contact HAVE THEM START READING THE SCRIPT That was great GIVE OUT TOKENS Ted where were you on the Feeling Thermometer during the role play OBTAIN A RESPONSE Also Ted what did you do that you liked and what one thing would you do differently OBTAIN A RESPONSE Abe where were you on the Feeling Thermometer during the role play OBTAIN A RESPONSE Abe what did you do that you liked and what one aspect would you do differently OBTAIN A RESPONSE Now let39s hear from the observers Tell us what you observed what you liked and one thing you would have done differently if you had been playing the role Also focus on ways that men are supposed to act that showed up here ENCOURAGE DISCUSSION Can you see how the ways in which males and females are supposed to behave is a different issue than whether you like someone of the sam esex You can fit the masculine stereotype and be gay or the feminine stereotype and be a lesbian Roles within a relationship can be negotiated as well Nowadays male and female roles are changing in ways that may seem confusing Fathers get maternity leave and stay home to take care of their children Women are top executives Gay and lesbian couples adopt children A lesbian partner is given guardianship over her comatose lover Yes there are stereotypes you have to deal with but you can decide how much you want to buy into them Exercise 6 Accepting Oneself In closing this session each person will receive a certificate for being who he or she is All you have to do to receive your certificate is to stand up and tell us why you are proud to be who you are Remember at the beginning of today39s session you came up with a list of why it is an advantage to be gay or lesbian Let39s take a look at that list again GO OVER THE LIST OF ADVANTAGES First I am going to pass out a blank card and pencil to everyone The card is for your notes Write what you plan to say to us on the card You have five minutes to make your notes on what you plan to say PASS OUT CARDS AND PENCILS ALLOW FIVE MINUTES CREATE A SEMICIRCLE EACH SPEAKER WILL COME TO THE OPENING lN THE SEMICIRCLE AS THE GROUP LEADER CALLS HISHER NAME AFTER THE PERSON HAS MADE THEIR REMARKS READ THE CERTIFICATE AND GIVE IT TO HERHIM CERTIFICATE THIS CERTIFICATE IS AWARDED TO FOR BEING WHO SHE HE IS Signature Date When I call your name please come to the opening in the semicircle and speak to us about why you are proud to be who you are Then I will give you your certificate BEGIN THE PROCESS AND APPLAUD AFTER EACH PERSON HAS RECEIVED THE CERTIFICATE Thank you Your speeches were wonderful For homework discuss why it is great to be gay or lesbian with a gay or lesbian friend Before we end today39s session I would like to know how your Feeling Thermometer readings are OBTAIN RESPONSES Finally take a few minutes and express appreciation to each person for their contributions and presence today ENCOURAGE GIVING STROKES TO EVERYONE I will see you next time at time on date END OF SESSION 3 CONFUSION CARDS JAHM CLNV DNDINIHL SI NOSHEId SIHl AVM HHl HlI A HERIDVSIG HO HERIDV HOA OC 39EEW HlI A DNOEU A DNIHLEWOS EH lelW HHEIHL 39GEISHOHV ATIVHXHS ERIHM AHHL NEIHM l OEW HWl STEP lHDIVHlS lVH A EDHT DNIHLANV CLNHOS LNOG AHHL 39quotRHD HEIHLONV HOH SDNITEIEM WHXHS AW HEHHOSEIG lNVO In GOOJ NVITVH SEDIH OHM NOSHEId HHl Ol SSVd NOISHJNOO S 3 CMVO JAHM CLNV DNDINIHL SI NOSHEId SIHl AVM HHl HlI A HERIDVSIG HO HERIDV HOA OC 39AVD EH LNVO IOS 39lON W I 39AHSIMS ERIV SAVD H18 39SAHD HEIHLO Ol GHlOVHllV THEM I CLNV ELLEITHLV OHOVW AHHA V W In GOOJ HSELNIHO SEDIH OHM NOSHEId HHl Ol SSVd NOISHJNOO S I CMVO CARD 3 S CONFUSION PASS TO THE PERSON WHO LIKES KOREAN FOOD quotMY NAME IS CLAUDE AND I REALLY LIKE EDWARD BUT TO SAY I39M quotFALLING IN LOVE SOUNDS WEIRD I GUESS GAYS CAN39T HAVE THE SAME FEELINGS AS STRAIGHTS DO DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY CARD 4 S CONFUSION PASS TO THE PERSON WHO LIKES GREEK FOOD quotI39M A VERY PRETTY AND FEMININE FIFTEEN YEAR OLD GIRL AND I FEEL SEXUALLY ATTRACTED TO OTHER GIRLS BUT I HATE TO DRESS LIKE A MAN OR ACT AGGRESSIVE SO I CAN39T BE A LESBIAN4AN I DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY CARD 5 S CONFUSION PASS TO THE PERSON WHO LIKES SOUL FOOD quotWHEN I WAS 14 I TOLD MY MOM THAT OTHER BOYS TURNED ME ON SHE SAID IT WAS JUST A PHASE IWOULD GROW OUT OF IT I39M 17 NOW NOT TO WORRY NEXT YEAR SHOULD BE THE LAST YEAR OF THIS PHASE DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY CARD 6 S CONFUSION PASS TO THE PERSON WHO LIKES LEBANESE FOOD quotMY NAME IS KEVIN AND I REALLY WANT THIS BOY IN MY ENGLISH CLASS IF I APPROACH HIM HE WILL REJECT ME AND TELL OTHERS I GUESS I CAN39T WANT HIM THAT IIUCH AFTER ALLquot DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY CARD 7 S CONFUSION PASS TO THE PERSON WHO LIKES YANKEE FOOD quotTHERE39S A GIRL IN MY GRADE WHOM EVERY ONE CALLS A LEZZIE KIDS ARE MEAN TO HER I THINK SHE39S EXCITING BUT IF I GOT TO KNOW HER THEY39D BE CRUEL TO ME TOO IT39S CONFUSING TO THINK THAT LOVE WILL BRING HATRED DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY CARD 8 S CONFUSION PASS TO THE PERSON WHO LIKES CAJ39UN FOOD quotWHEN I WAS A KID THIS FRIEND OF THE FAMILY MADE ME TAKE HIS PENIS IN MY MOUTH THAT39S WHY I39M GAY DO YOU AGREE OR DISAGREE WITH THE WAY THIS PERSON IS THINKING AND WHY JAHM CLNV DNDINIHI SI NOSHHCI SIHI AVM EIHI HIIM EIERIDVSIG HO EIERIDV IIOA OC 39HVEIA IXEN NIVDV II AHI TLI IIIEI LEIEIH EDIIT LNGIG I 39HIHIIISOHd V OI OD OI 3W CquotIOI EIH 39IAIEIHI gtIOI1S LNOG lelf AEEHI 39SEISINEICI SANEEW HHHIO IIIOEIV gtINIHI NEW TIV CEIVS EIH 39AVD SVM I IHDIIOHI I CVC AW CquotIOI In GOOJ NVOIXHW SEDIIT OHM NOSHEICI EIHI OI SSVcI NOISIIJNOO S 01 CIEIVO JAHM CLNV DNDINIHI SI NOSHHCI SIHI AVM EIHI HIIM EIERIDVSIG HO EIERIDV IIOA OC NVIEISEIT V EEWOOEIEI OI IIIEI EDIOHC ON CIVH I IVHI HELIHV 39MONH LNGIG I 3W IIIOEIV DNIHIEWOS GEISNEIS EIAVH ISM EIHS 39II EIAEHTEIEI LNVO I 39GHSHOHV OS 109 I 39NEIAEITEI SVM INEIHM EW GHOIIGEIS NVIEISEIT HHGTO SHIIH SEHIOEIERIVEI SEDIIT OHM NOSHHCI EIHI OI SSVcI NOISIIJNOC S 6 GHVO NELS 18 1988 426 Reformatted in 2002 and 2008 preserving original page numbering for the most part but di ering in the use of fonts including Arborwin font for trees BINOMINAL EACH KEN SAFIR and TIM STOWELL Rutgers New Brunswick and UCLA 10 A Dyadic Quanti er Within the research tradition of generative grammar quantifiers have typically been assumed to be fundamentally different from predicates such as verbs or adjectives insofar as only the latter categories take one or more arguments to which they assign grammatical functions such as subject object etc Thus in a sentence like la the predicate loves assigns its thematic roles to a subject every man and an object Jane The LF representation lb treats the subject of loves as a variable bound by the quantifier Crucially the quantifier introduces no new grammatical relation of its own 1 a Every man loves Jane b every mani ei loves Jane 427 BINOMINAL EACH We contend that there is at least one natural language quantifier that acts exactly like adjectival or verbal predicates in that it has a dyadic argument structure parallel to that of an adjective The quantifier in question is each in a particular construction that we will call binominal each This is exemplified in 2 where each distributes over two NP arguments 2 The men saw two women each A number of issues about the nature of LF operations will emerge from our analysis of binominal each but we reserve these matters for their natural place in our presentation 11 Some Relational Properties of Binominal Each The structural position each in 2 is not immediately apparent in that each might be either a subconstituent of the direct object NP or a direct constituent of the VP analogous to the structure in 3 where it occurs in VP initial position 3 The men have each seen two women We will refer to this usage of each as adverbial each to distinguish it from binominal each in 2 Two factors argue in favor of an NPinternal position for binominal each First if the VP does not contain a direct object then each may not occur to the right of the verb 4a The men each decided to leave b The men decided to leave each The contrast between 4a and 4b suggests that true adverbial each may only occur VPinitially and that 2 involves a distinct structure with each as a subconstituent of NP This is confirmed by the paradigm in 56 where the direct object undergoes movement cf Burzio 1981 1986 5a How many girls each did the men see b One girl each was seen by the men 6a How many girls did the men see each b One girl was seen by the men each c One girl was seen each by the men d How many girls did the men each see KEN SAF IR AND TIM STOWELL 428 The contrast between 5 and 6a c follows automatically if binominal each is a subconstituent of NP in 2 When each does not occur as a subconstituent of NP it must occur in VPinitial position as in the adverbial each construction in 3 4a and 6c The contrasts between these two usages of each extend to a number of other phenomena that lie beyond our immediate concerns By distinguishing binominal each from adverbial each we do not intend to imply that there are two distinct homophonous leXical items each Rather we suggest that there are two clusters of properties that each has depending on its syntactic position and its interpretation In the remainder of our discussion we confine our analysis to binominal each Binominal each constructions impose certain restrictions on the two NPs that each takes scope over We will refer to the NP containing postnominal each as the Distributing NP DNP thus in 2 two women is the DNP The DNP must always be cardinal and indefinite definite NPs bare plural NPs and quantified plural NPs are all excluded 7a The men saw one jewel each b The men saw a jewel each c The men saw two several jewels each d The men saw somecertain jewels each e The men saw thethose jewels each f The men saw bothmosUall jewels each Note the contrast between cardinal indefinites 7a c and noncardinal indefinites 7b d Many speakers find singular indefinite D NPs like that in 7b fully acceptable perhaps indicating that the indefinite article aan can function as a numeral in this dialect Among DNPs of the form X of the Ns each we find the judgments in 8 8a The men saw onetwo several of the women each b The men saw somemanyfew of the women each c The men saw mosUallboth of the women each The other NP which we refer to as the Range NP RNP is typically plural and specific it corresponds to the NP the men in 2 429 BINOMINAL EACH The range of possible RNPs is illustrated in 9 9a TheyThe menThose menThe five men saw two women each b Bill and Joe saw two women each cSome men Several menMany men saw two women each d Five menA few menA group of men saw two women each eThe manA man Someone SheJoe saw two women each EveryoneEvery man saw two women each g All the menBoth the men saw two women each h All menBoth menMost men will see two women each i TwoManySeveralA lot of the men saw two women each j Martian men marry two women each k No menNo manFew men married two women each The RNP may be a de nite plural 9a or a conjoined definite NP 9b As 9cd show the RNP may also be an indefinite plural NP but the interpretation of 9cd clearly requires a specific reading of the indefinite RNP 9e indicates that the R NP may not be a singular NP regardless of whether it is definite indefinite quantified or whatever Although everyone is marginally possible as an RNP in 9f it is probably a leXical idiosyncrasy of everyone that it behaves like a plural since every man is plainly worse39 cf Williams 1986 When the RNP is a universally quantified plural NP as in 9gh or a partitive NP as in 9i the judgments are delicate but the sentences seem basically acceptable and generic plurals 9 seem fine Negatively quantified plurals 9k are excluded as RNPs In this presentation we shall not attempt to devise a single characterization that will pick out all possible DNPs or all possible RNPs but these distinctions serve to illustrate the asymmetry between the arguments of binominal each and will play essentially a diagnostic role in our analysis 12 The Interpretation of Binominal Each Our terminological distinction between the DNP and the RNP is based on the logical interpretation of these constructions In 2 binominal each effects a mapping between individual men and sets of two women such that the men see the women Generalizing we suggest that 10 provides a rough informal characterization of the interpretation of binominal each KEN SAF IR AND TIM STOWELL 430 10 The individuals in the set denoted by the RNP are exhaustively mapped onto sets denoted by the D NP such that no two Rindividuals are mapped onto the same D set Thus in a sentence like 1 l at least siX books must be purchased and no two of the men can have combined in the purchase of any of the books We owe the latter observation to James Higginbotham 11 Three men bought two books each This may eXplain why the cardinal NPs in 8a are awkward since the exhaustive mapping is limited to a portion of the individuals in the larger set of men Insofar as it establishes a relation speci cally between two NPs binominal each is unlike most other quantifiers It is instructive to show how binominal each differs from other floated usages of each such as adverbial each which relates the subject and a VP These two usages are contrasted in 12 12 a The girls each had a good time b The girls had a good time each c The girls met a boy each Abstracting away from the marginality of the noncardinal indefinite DNP in 12c we note that 12b fails because have a good time is an idiom where a good time fails to refer and cannot serve as a cardinal DNP 12a is grammatical because adverbial each is directly adjoined to VP and does not require a DNP as shown above 20 The Syntactic Diathesis 0f Binominal each 21 Partitive Each As is well known each also occurs prenominally either as a specifier or a singular NP as in 13a or as the specifier of a partitive NP as in 13b 13a Each boy boys went home b Each of the boys boy went home The olePs that may follow each in a partitive NP are a proper subset of the class of RNPs selected by binominal each in 9 431 BINOMINAL EACH 14 a Each of them left b Each of thethoseMo sthe ten men left c Each of fivea few men left d Each of Bill and Joe left e Each of someseveralmanya few men left f Each of all the menboth the men left g Each of allbothmost men are tall h Each of thethat a every man left i Each of Martian men have two antennae j Each of nofew men left With the exception of conjoined names and quantified NPs the set of possible R NPs is equivalent to the set of possible ofNPs occurring with partitive each We suggest that partitive each should be analyzed as the specifier of an NP headed by a null singular cardinal noun or pronoun parallel to the oven cardinal proform one in 15 15 a Each one of them left b Each one of thethoseMo sthe ten men left Each one of fivea few men left Each one of Bill and Joe left Each one of someseveralmanya few men left Each one of all the menboth the Men left g Each of allbothmost men are tall h Each of nofew men left C9338 Although the questionable examples 15df are somewhat better than their counterparts in 14 the pattern of judgments is basically similar This suggests 17 as the structure for partitive NPs with prenominal each in 14 where each occurs in the Specifier position of NP and the null head N acts as a proxy complementtaker 16 NP each N N 911 9pr AltematiVely we might analyze each and other quantifiers as heads of QP taking N complements in the spirit of Abney 1986 Unlike most quantifiers each has the added ability to sanction an empty N head KEN SAF IR AND TIM STOWELL 432 22 A Structure for Binominal Each Returning now to binominal each phrases their structure might be similar to that of partitive each One obvious difference between the two constructions is that no overt material follows each in the binominal examples so the oleP following the null head N in l 6 would itself have to be null with binominal each 17 NP NP NP two books eachi N N eli NP 6 This empty NP could then be treated as a kind of anaphor taking the RNP as its antecedent thus providing the basis for a possible account of the rough correlation between the class of possible RNPs in binominal each constructions in 9 and the class of possible olePs in the partitive constructions in 14 and 15 Alternatively the binominal each phrase might have the structure in 18 if the quantifier is really the head of QP rather than a Specifier of NP 18 NP NP QP two books SPEC e N each li NP e An obvious objection to 17 or 18 as the structure of the D NP is that it fails to eXplain the fact that neither an overt one nor an overt oleP may follow binominal each 19 a The boys bought two books each one of them b Sam and Bill saw two women each e of them c How many books each one of them did the boys see d How many books each e of them did the boys see 43 3 BINOlVHNAL EACH We believe that the structural parallel between binominal each and partitive each is genuine and that the structures in 19 are excluded on Casetheoretic grounds However we shall not develop this analysis here for the sake of brevity Whether there is evidence favoring 17 over 18 will not be examined here Rather we wish to exploit a key property that these analyses have in common in both 17 and 18 there is an empty object complement following each First we suggest that the presence of the null NP provides an explanation for the postnominal position of each within the D NP Second we suggest that the anaphoric relation holding between the null NP and its antecedent the RNP provides the basis for an explanation of certain restrictions on binominal each constructions The rest of the paper develops these arguments more fully 23 AParallel with AP Turning first to the issue of why binominal each phrases occur post nominally within the D NP we suggest that this mirrors the distribution of adjectival modifiers in NP It is well known that modifying adjectives may appear postnominally only if they take complements39 otherwise they appear pre nominally 20 a A man happy about his plans discussed his hopes b A student willing to try is likely to succeed c A man happy discussed his hopes d A student willing is likely to succeed 21 a A happy man an obvious fact two crafty cooks b A happy about his plans man discussed his hopes c A willing to try student is likely to succeed The prenominal APs in 21bc can be excluded by Williams s 1982 Head Final Filter while the postnominal APs in 20cd can be excluded under the assumption that an adjective with no internal argument must adjoin to the left of a head noun perhaps by an incorporation rule of the sort suggested in Stowell 1981 cf Baker 1 985 KEN SAF IR AND TIM STOWELL 434 As observed by Safir 1985 a few adjectives such as present must appear post nominally despite having no overt complement 22 a One man presenUsick complained about the food b One sickpresent man complained about the food Such adjectives can be naturally analyzed as unaccusatives in the sense of Perlmutter 1978 ie as ergatives in Burzio s 1981 terminology On this view these adjectives are monadic predicates selecting a single internal argument If this null argument is equated with PRO it presumably undergoes movement to the ungoverned Subject or Spec position in AP cf Stowell 1983 where it may be controlled by the head NP 23 NP NP AP one man Spec A PROi present t i If present always requires an empty category object its inability to appear pre nominally can be attributed to Williams s 1982 Head Final Filter The empty object would thus eXplain the postnominal position of present in 221 Binominal each occupies the same postnominal position as the adjectives just described This suggests that each likewise has a null complement when it occurs postnominally within NP as we have already suggested The existence of a null object would thus reduce the postnominal position of the binominal each phrase to the same factor determining the position of AP modifiers Summarizing the eachphrase is adjoined to the D NP because the D NP controls the eachphrase s PRO subject just as NP controls the PRO subject of an AP or PP modifier or the Wh pronoun in a relative clause The eachphrase is rightadjoined to the D NP because of its null internal object argument which invokes a Headfinal Filter effect preventing the EP from occurring prenominally in the D NP 43 5 BINOMINAL EACH 30 Structural Constraints on the RNP DNP Relation 31 Background Assumptions Assuming that we are correct in supposing that binominal each governs a null object complement the question arises whether each is a monadic unaccusative predicate like present or a dyadic predicate like aware lf binominal each is monadic we would expect its null object to be a trace bound by PRO in the Spec position of AP as in 23 On the other hand if binominal each is dyadic then its structure would more closely resemble 24 where the subject argument of each is a PRO controlled by the D NP head and the null object argument of each is a different type of empty category one that is not bound within the D NP 24 NP NP QP two books Spec Q PRO i each e j We suggested above that the null object of binominal each is anaphorically related to the R NP in some way So far we have not considered the nature of this binding relation in any detail but if the null object is indeed a type of anaphor that requires a plural R NP antecedent we would expect the binding relation between them to affect their relative structural positions This would inevitably affect the structural position of the D NP as well since the EP is adjoined to the D NP In this respect our theory recalls Burzio s 1981 1986 account of RNP and D NP distribution in terms of the idea that binominal each is itself an anaphor2 In this section we examine the syntactic distribution of the RNP and the D NP There are several interesting distributional restrictions exhibited by these NPs which we will use as clues to the nature of the binding relation involved in this construction As we shall see the observable restrictions on the distribution of the RNP and D NP can be interpreted theoretically in many ways by invoking KEN SAF IR AND TIM STOWELL 436 various combinations of familiar structural conditions such as ccommand subjacency CED Binding Condition A SSCNIC etc 32 The DNP May Not Be a Subject In simplex sentences with active transitive verbs the RNP may occur as an external argument in the subject position of IP but the D NP may not 25 a The men saw one woman each b John and Bill gave presents to one woman each c Sam and Dave will love one woman each 26 a One student each left b One woman each saw John and Bill c One student each gave presents to the teachers d One woman each loves Sam and Dave e One student each received the presents f One woman each will please Sam and Dave In all of the examples in 26 the subject of IP is singular thus ruling out the adverbial each interpretation These examples show that D NP subjects are uniformly excluded Notice moreover that grammatical function rather than thematic role is involved here The D NP may occur as the Goal indirect object of give in 25b and as the Theme object of love in 25c but not as the Goal subject of receive in 26e or as the Theme subject of the Psychverb please in 26f3 There are three ways of interpreting the prohibition against the D NP occurring as an external argument in the subject position of a simplex clause First one might assume that the RNP must itself occur in the subject position for some reason thereby preventing the D NP from occurring there We consider this possibility in Section 33 Second the structural relation between the RNP and the DNP might be the crucial factor For example it is possible that the R NP must ccommand either the D NP or some element within it such as each or its null object We will examine Burzio s proposals along these lines in Sections 34 and 35 43 7 BINOlVHNAL EACH Third one might assume that the D NP simply can t occur in a subject position of any sort We will cite evidence supporting this view in Section 35 and in Section 4 we will provide a possible motivating principle for this 33 The RNP May Be an Object The first idea that the RNP must be a subject can be dismissed on the basis of examples such as the following 27 a Jo sentintroduced his kids to two coaches each b The dean put the professors on one committee each c Mary sent one book each to the professors d The capt presented to his spies five medals each e Mary sent the professors one book each 28 a One coach each senUintroduced his kids to Bill b One dean each put the professors on the board c One woman each sent the books to the professors d One capt each presented to the spies the medals e One woman each sent the professors the books In all of these examples the subjects are singular so the RNPs must be either direct or indirect objects In 28 the RNPs occur in the same positions as in 27 Evidently the ungrammaticality of D NP subjects can t be attributed to any general prohibition against VP internal RNPs 34 Reconstruction CCommand and Principle A Recall that D NP objects are unaffected by preposing under Wh movement in 5a b despite the reversal of the relevant precedence and ccommand relations The toomovement construction in 38b exhibits the same phenomenon as do Burzio s oftcited examples involving Passive Raising and Pseudocleft constructions in 3 9 29 a How many women each do you think the boys visited b Five books each is too much for the boys to read 30 a One interpreter each was assigned to the visiting diplomats b One book each appears to have been given to the boys c One interpreter each is what they want to have All of the examples in 29 and 30 are grammatical despite the fact KEN SAF IR AND TIM STOWELL 438 that the D NP precedes and asymmetrically ccommands the RNP in every case As 29b and 30 show the D NP subject prohibition does not necessarily hold if the subject position in question is a nontheta position Evidently the structural condition responsible for 2628 must hold either at Dstructure as in Burzio 1981 or at LF as in Burzio 1986 As Burzio remarks the strongest evidence for an LFbased account comes from examples like 30c where the RNP neither precedes nor ccommands the D NP at either Dstructure or S structure He notes that judgments about quantifier scope provide independent evidence for the possibility of reconstructing pseudo clefted constituents to the position of Wh trace ie to the position occupied by Wh at D structure Assuming that this option is also available for the D NPs in 29 30 the most natural way of accounting for the ungrammaticality of 2628 is to assume that the relevant structural condition holds at LF Burzio 1986 maintains that the LF ccommand relation between the R NP and the D NP is the crucial factor in 2628 He seeks to derive this from Principle A of Chomsky s 1981 Binding Theory under the assumption that binominal each is an anaphor with the R NP as its antecedent Since the RNP must bind each in order to satisfy Principle A it follows that it must ccommand the D NP containing each at the level where the Binding Theory applies The subject asymmetrically ccommands all VP internal arguments so Principle A is violated in 26 and 28 It would also be violated in 2930 if the Binding Theory were assumed to hold at S structure since 2930 are grammatical he concludes that the Binding Theory applies to the output of Reconstruction at LF cf Chomsky 1981 145 Belletti and Rizzi 1986 Burzio provides independent support for the relevance of Principle A to binominal each constructions He shows that it is not sufficient for the RNP to c command the D NP and thereby bind each in addition the RNP binder of each must occur within the governing category of each He cites examples like the following as classical NIC and SSC effects 43 9 BINOlVHNAL EACH 31 a The boys said that three women each had left b The boys expected Mary to kiss one child each In each case the RNP falls outside of the governing category of the D NP and the sentences are ungrammatical Thus there seems to be some empirical justification for Burzio s bindingtheoretic account 35 Problems With the Condition A Based Account We see two fundamental problems with Burzio s proposal First as Burzio acknowledges the D NP may not occur in a thetamarked subject position even when Principle A would be satisfied 32 a The boys expected that pictures of each other would be on sale b The boys expected pictures of themselves to be on sale c The boys considered themselveseach other to be smart d The boys believed that themselveseach other were smart 33 a The boys expected that one picture each would be on sale b The boys expected one picture each to be on sale c The boys considered one girl each intelligent If each is the relevant anaphor as Burzio suggests then all of the examples in 33 should be grammatical assuming that each has the same governing category as the gardenvariety anaphors in 32 Even if one assumes that the LGB accessibility condition does not apply in 33 so as to exclude 33a on par with 32d this would still fail to account for the exclusion of 33bc where an NlCstyle account is unavailable Similar problems arise if the relevant anaphor is assumed to be either the null object of each or the entire D NP If the former 33a is allowed unless the accessibility condition is dropped and 33bc are permitted regardless If the latter 33a is excluded but 33bc are again permitted Thus some additional principle is needed to exclude some or all of the examples in 33 and it is possible that this principle would also account for the exclusion of 2628 KEN SAF IR AND TIM STOWELL 440 A second problem with Burzio s bindingtheoretic account lies in an empirical claim underlying it namely that the RNP must ccommand the D NP in its reconstructed position at LF He observes that indirect object NPs are free to serve as RNPs as in examples like 34 34 a The UN assigned one interpreter each to the Visiting diplomats b John talked to Sam and Tom about two women each Burzio suggests that the indirect object must be able to ccommand other constituents of VP This would follow if the tophrase is an NP projection of the indirect object rather than a PP projection of to Although Burzio claims that other types of PPs may not harbor RNPs we find that this is possible in many cases4 35 a Tom is depending on the boys for two ideas each b Mat lived with Sue and Mo in one apartment each c Mat worked with Sue and Mo on two projects each d Reagan tried to put one medal each on the spies e John blamed three crimes each on the prisoners In some cases it even appears to be possible for the R NP to be embedded in a small clause or ECM clause with the D NP in a matriX adjunct phrase5 36 a Jones proved the prisoners guilty with one accusation each b Bob madelet Sam and Tom leave on two occasions each The distribution of true re exive and reciprocal anaphors suggests that the RNPs in 3536 do not ccommand the D NP positions in question 37 a Tom lived with Sue and Mo in each others apartments b Reagan tried to put themselves on top of the spies c John blamed each others antics on the prisoners d J ones proved the men guilty with each others confessions We conclude that the RNP needn t necessarily ccommand the D NP 441 BINOlVHNAL EACH even at LF since none of the examples in 3536 contain a trace position within the ccommand domain of the RNP that the D NP could reconstruct into 36 Summary We have seen that some of the structural constraints on the distribution of the D NP and RNP lend prima facie support to a binding theoretic account of binominal each relying on Principle A However this account is probably too strong in requiring that the RNP must ccommand the D NP and too weak in failing to exclude structures involving D NPs in subordinate subject positions In the next section we will propose an alternative theory of some of these effects which we believe provides a more satisfactory account of the structural constraints on the RNP and D NP discussed here 40 The LF analysis The theory of binominal each must capture three basic generalizations which we state in 38 38 a The D NP may not be a D structure subject b The structural relationship between the D NP and RNP is clausebound in that the D NP may not occur in a more deeply embedded clause than the R NP although the reverse is some times possible with nonfinite clauses c The structural relation between the RNP and D NP exhibits reconstruction connectivity effects We will account for these as follows Regarding 3 8a we suggest that the subject prohibition is a special case of Chomsky s 1973 Subject Condition on movement More specifically we suggest that the each phrase EP undergoes movement out of the D NP at LF and adjoins to IP Regarding 38c we assume that the D NP may reconstruct into any trace position bound by it including its D structure position and that the LF movement of the EP may originate from the reconstructed position thereby evading potential Subject Condition violations in some instances Before turning to the problem posed by 38b we will develop this analysis in a bit more detail and then provide some independent evidence in support of it KEN SAF IR AND TIM STOWELL 442 41 The Subject Condition The fact that the D NP must be a D structure object or indirect object suggests that either the D NP or some element within it must undergo movement at LF If the entire D NP undergoes LF movement then we might interpret the subjchobject asymmetry in D NP distribution as a classical ECP style effect under the assumption that the trace of the D NP is not properly governed Alternatively if the eachphrase or the null object of each must undergo movement then the subjchobject asymmetry might really be a Subject Condition effect as suggested above The fact that the D NP may not be the subject of a small clause or ECM infinitival clause in 33bc shows that the Subject Condition rather than the ECP is at work here If the entire D NP were to undergo LF movement in these examples then no ECP effect should arise since LF extraction of Wh phrases and other QPs is perfectly grammatical in these contexts 39 a Someone considers everyone to be foolish b Who believes who to be foolish c The boys consider one girl each to be foolish On the other hand if the eachphrase or the null object of each undergoes LF movement then 33bc and 39c are correctly excluded as Subject Condition violations parallel to structures involving overt movement 40 a Who did you buy a picture of b Who do you believe a sister of to have left c Of which book do you consider a review important Suppose then that binominal each constructions involve LF movement out of the D NP For concreteness we will assume that the null eachphrase QP undergoes movement and adjoins to IP 5 just like other nonWh QPs Most theories of movement derive Subject Condition effects from other more general principles Kayne 1983 derives them from the Connectedness Condition Huang 1982 derives them from the Condition on Extraction Domains CED 443 BINOlVHNAL EACH and Chomsky 1973 1981 1986 derives them from Subjacency For the purposes of our presentation it doesn t really matter which of these theories is adopted as long as the effect holds at the level of LF It is however incumbent upon us to justify the obligatory nature of this LF movement so as to force the SubjacencyCED effect 42 LF Movement of Binominal Each Suppose that the null object of each must be locally Abar bound in order to be licensed as a variable at LF The object argument of each is selected to be an R NP and so the R NP must be the A binder at least by LF Suppose further that there is a locality restriction on this binding relation in the spirit of Aoun 1985 such that the variable must be A bound in its governing category The latter stipulation will force the eachphrase to move out of its D NP since the D NP will be or will contain the governing category for the empty category and so the latter cannot be A bound by the RNP unless the eachphrase escapes the D NP It follows that the Subject Condition will then be violated whenever the D NP is in subject position because the eachphrase must always be extracted from the D NP But how then does the RNP come to be a local A binder We have assumed as is generally the case for QR that the eachphrase adjoins to IP If the RNP is a Whphrase in Comp as in 41ab it will locally Abar bind the null object directly39 otherwise the RNP will also have to undergo QR in order to Abar bind the null object of each as in 42cd 41 a Which men bought one book each b Which men1 each e1 2 t1 bought one book t2 42 a The men saw two women each b The men1 each e1 2 t1 saw two women t2 This LF derivation correctly predicts the relative scope relation between the RNP and D NP KEN SAF IR AND TIM STOWELL 444 43 a Fifty men each saw two women b Fifty men saw two women each 43a is ambiguous in that it allows either cardinal QP to take broad scope39 but 43b requires the cardinal D NP to take narrow scope with respect to the RNP subject This is expected given our LF representation of binominal each The R NP must take scope over the eachphrase in order to bind the null object of each and the eachphrase in turn must take scope over the D NP in order to bind its own trace within the D NP 43 Reconstruction Effects Now consider the reconstruction effects noted in 38c Assuming with Chomsky 1981 and Burzio 1986 among others that Reconstruction is freely available at LF the D NP may return to its Dstructure position This explains the fact that the D NP is free to undergo syntactic Amovement to a subject position as in 30ab above If the D NP reconstructs to an object position the eachphrase is then free to move out of the D NP without incurring a Subject Condition Subj acency violation Thus 44a would have the LF structure 44b 44 a Two interpreters eachi seem ti to have been ti assigned ti to the diplomats b ei seem ei to have been ti assigned two interpreters eachi to the diplomats In terms of Lasnik and Saito 1984 and Chomsky 1986 this implies that the object position retains its gamma feature after the D NP has reconstructed into it thus ensuring that the D NP does not act as a Subj acency barrier This derivation is not available to nonderived subject D NPs since they have no direct object trace position to reconstruct into Our account predicts that not all derived subjects can be legitimized in this way If the trace position that the D NP reconstructs into is itself a subject position then reconstruction is of no help in avoiding a Subject Condition violation Thus examples like 45 are correctly excluded 445 BINOlVHNAL EACH 45 a Two women each seemed to the men e to have shot themselves b Two women each seemed e to love the men Finally consider the status of reconstruction effects with D NPs involving Abar movement as in 29 30c and 46a 46 a How many books eachi did the men say ti the boys read ti b e did the men say the boys read how many books eachi c e did the men say how many books eachi the boys read ti In each case the D NP can reconstruct to an object position as illustrated by 46b The eachphrase is free to move out of the D NP without incurring a Subj acencyCED violation as before In principle the LF derivation in 46c is also permitted since the D NP is free to reconstruct to the position of an intermediate trace left by successive cyclic Wh movement Evidence for this sort of reconstruction has been cited by Barss 1986 and Williams 1986 with respect to the binding of re exive pronouns 47 a John said that Mary bought a picture of himself b Which picture of himselfi did John say ti Mary bought ti The fact that the re exive pronoun may be bound by John in 47b but not in 47a suggests that the Wh phrase or a subconstituent thereof can reconstruct to the intermediate trace position in the Spec of the embedded CP where it can be locally Abound by John But the analogous interpretation with binominal each is completely excluded Thus in 46a the only possible R NP is the embedded subject the boys the matrix subject the men cannot serve as the RNP Our theory predicts this since the intermediate trace position is not a thetamarked object position and so subsequent extraction of the eachphrase out of the D NP would incur a KEN SAF IR AND TIM STOWELL 446 Subj acencyCED Violation analogous to extraction from the reconstructed subject positions in 45 44 The Locality ofEachmovement Our analysis accounts for the fact that the D NP may not be a D structure subject regardless of whether it is ccommanded by the RNP If the RNP fails to ccommand the D NP as in 3 536 it is still free to undergo LF movement into a position where it may locally Abar bind the null object of each However we still encounter an important problem our analysis does not exclude the possibility of the eachphrase undergoing successivecyclic movement to a higher clause where it might be bound by a distant RNP as in 48 48 The boys said Mary captured two snakes each Notice that we can t appeal to Condition A of the binding theory to force the R NP to occur within the governing category of the D NP in light of 3 536 It seems that the only option available is to assume that the binominal each phrase is unable to undergo successive cyclic movement for some reason A possible explanation for this concerns the dyadic argument structure of binominal each Recall that the eachphrase is a modifier of the D NP and contains a PRO subject argument bound by the head of the D NP Since extraposition of NP modifiers is in general clausebounded we must assume that some principle blocks successive cyclic movement of modifiers39 see Gueron and May 1984 for discussion of this It seems reasonable to suppose that the same principle is at work in constraining LF movement of the binominal eachphrase The apparent clause mate restriction holding between the RNP and the D NP will then follow from the locality of eachmovement the only exceptions occurring in structures like 36 where the RNP is evidently permitted to QR out of its immediate clause at LF 50 Concluding Remarks Our analysis of binominal each has touched on a number of general issues all of which ultimately deserve a deeper treatment 447 BINOlVHNAL EACH First our analysis suggests that quantifiers may have a dyadic argument structure just like a conventional dyadic verbal or adjectival predicate Although most quantifiers any all some etc are monadic intransitives or perhaps unaccusatives the behavior of binominal each suggests that this is not a necessary property of quantifiers We expect that it will prove fruitful to compare the properties exhibited by binominal each with those of other natural language quantifiers that may be analyzed as syntactically and semantically dyadic One such quantifier is the resultative operator so which must govern its complement clause at LF according to Gueron and May 1984 as in John talked to so many people that he was exhausted The idea that the diathesis of so must be satisfied at LF is similar to our claim that the eachobj ect receives its content at LF Second our account of the locality conditions on binominal each constructions relies on the idea that LF movement is subject to the standard conditions on syntactic movement thus providing further support for proposals along these lines in the references cited above However the fact that adjuncts behave as weak islands with respect to LF extraction of each raises an interesting descriptive problem for this view eg The men cut the salami with one knife each Third our assumption that the RNP must also undergo QR in order to bind the null object of each implies that plurals including conjoined NPs such as John and Bill must be able to undergo QR For suggestions along these lines motivated by other concerns see Huang 198239 p 269ff and Clark forthcoming among others We believe that our analysis captures the core of the binominal each phenomenon and that it has a variety of interesting consequences However we suspect that we have only scratched the surface of many of the semantic and syntactic issues that may be examined by means of this construction KEN SAF IR AND TIM STOWELL 448 NOTES We are indebted to Jim Higginbotham for much helpful discussion and for written comments on the contents of a preliminary draft of this paper This is a somewhat abridged version of a longer paper in preparation on this topic 1 Another class of superficial exceptions to the generalization that postnominal adjuncts must have a complement turn out to prove the rule We have in mind verbal passives which may appear in postnominal position without an overt complement But verbal passives presumably have a post verbal trace and if so should act like present By contrast it is much more difficult to place adjectival unpassives in the same contexts i Some of the food touched was contaminated ii Some of the untouched food was contaminated iii Some of the food untouched was nonetheless contaminated 2 An alternative analysis for the internal argument of binominal each would be to assume that the empty category arises by Abar movement perhaps of an empty operator At present we can see no advantage to such an account and so we will not explore this possibility 3 The exclusion of 26f provides evidence against the analysis of Psych predicates proposed by Belletti and Rizzi 1986 where it is claimed that these subjects originate in direct object position at D structure The subsequent text discussion of reconstruction effects is of direct relevance in this respect 4 We do not have a well worked out account of the contrast between Burzio s PP data and the examples in 35 5 Although the data are too complex for us to discuss them here we believe that dative structures allow for either the dative or the direct object to be the RNP or D NP once a number of peculiar restrictions are controlled for We hope to treat this issue in a lengthier treatment of these issues 449 BINOIVIINAL EACH 6 The distribution of Wh in situ led Chomsky 1973 and Huang 1982 to assume that Subjacency and CED do not apply at LF While Subject Condition effects are more robust than other island effects at LF cf Kayne 1983 the absence of other island effects suggests the neutralization of Subjacency This would create a possible problem for our account we rely on the idea that Subjacency constrains the movement of the eachphrase out of the D NP at LF Recent work by Lee 1982 Pesetsky 1987 and Nishigauchi 1984 however suggests that Subj acency does hold at LF 7 Other candidates for dyadic quantifiers might include comparatiyes or and perhaps polyadic and BIBLIOGRAPHY Abney S 1986 The English Noun Phrase in its Sentential Aspect Unpublished MIT PhD dissertation Aoun J 1 985 A Grammar of Anaphora MIT Press Cambridge Baker M 1985 Incorporation A Theory of Grammatical Function Changing Unpublished MIT PhD dissertation Barss A 1986 Chains and Anaphoric Dependence Reconstruction and its Implications Unpublished MIT PhD dissertation Belletti A and L Rizzi 1986 Psych Verbs and Thetatheory to appear in Natural Language and Linguistic Theory BurZio L 1981 Intransitive Verbs and Italian Auxiliaries Unpublished MIT PhD dissertation BurZio L 1986 Italian Syntax Reidel Dordrecht Chomsky N 1973 Conditions on Transformations in S Anderson and P Kiparsky eds A F estschrift for Morris Halle Holt Rinehart and Winston New York Chomsky N 1981 Lectures On Government and Binding Foris Dordrecht Chomsky N 1986 Barriers MIT Press Cambridge Gueron J and R May 1984 Extraposition and Logical Form Linguistic Inquiry 1139 637678 Huang 1982 Logical Relations in Chinese and the Theory of Grammar Unpublished MIT PhD dissertation Kayne R 1983 Connectedness Linguistic Inquiry 1439 223249 Prepublication version of Tim Stowell 2007 Sequence of Perfect Published Version Tim Stowell 2007 Sequence of Perfect in Louis de Saussure Jacques Moeschler and Genoveva Puskas eds Recent advances in the syntax and semantics of tense mood and aspect Trends in Linguistics Vol 185 Berlin Mouton De Gruyter pp 123 146 Sequence ofPerfect Tim Stowell UCLA 0 Introduction In this article I will argue that the English infinitival perfect haveen functions as a true past tense in at least some cases reviving an old analysis Hoffman 1966 I will review Hoffman s three main arguments for this position and reject two of them while accepting the third In addition I will show that the infinitival perfect resembles the English preterit past in exhibiting a simultaneous presenttenselike interpretation when embedded within a main clause containing past tense This type of interpretation a paradigmatic example of sequence of tense is commonly assumed to be possible only with finite tenses The broader implication is that in nitival clauses may contain tensesiat least past tense 1 Temporal argument structure and interpretation In a main clause tenses conventionally convey a temporal relation between the actual Utterance Time UT and what Klein 1994 calls the Topic Time In simple sentences lacking auxiliary verbs Klein s Topic Time TT corresponds roughly to the Reichenbach s 1947 traditional notion of the event time ET but in sentences containing aspectual auxiliary verbs the TT is a time related to the ET by the aspectual semantics of the auxiliary Following previous work I assume that tenses are dyadic predicates expressing a temporal ordering relation holding between two timedenoting arguments I refer to the external argument of a tense as its Reference Time argument RT the internal argument of the tense is Klein s TT In a main clause the RT of a tense denotes the actual UT Thus a main clause tense orders the UT in relation to the TT Past tense is assumed to be a temporal ordering predicate meaning after it orders its external RT argument denoting the UT after its internal argument the TT In lac the past tense locates the actual UT after the TT 1 a Max ate an apple b Max had eaten an apple c Max was eating an apple In la there is no aspectual auxiliary so TT and ET coincide thus the past tense orders the UT after the ET the time interval of the appleeating event In lb and 1c the periphrastic aspectual auxiliary constructions express a temporal ordering relation between the TT and the ET The periphrastic perfect haveen resembles the preterit past tense in expressing anteriority or pastshifting just as past locates UT after TT so haveen locates TT after ET The periphrastic progressive being locates the TT within the ET here the TT denotes a subinterval of ET Consequently lb and 1c involve reference to three distinct times UT TT and ET represented schematically in the traditional timeline diagrams in 2 where time ows from right to left 2 a UT 7 TT ET lb b UT 7 TT l 10 In subordinate clauses tenses work somewhat differently First they may be used to express a relation between the TT of the subordinate clause and a time other than the actual UT Typically the other time in question is the ET of the matrix clause as in 3 3 Bill said that Max ate an apple In 3 the main clause past locates the UT after the main clause TT Since there is no aspectual auxiliary in the main clause the main clause TT TTl is the main clause ET ET 1 denoting the time at which Bill spoke The subordinate complement clause also contains no aspectual auxiliary so its TT TT2 coincides with its ET ET2 denoting the time at which Bill ate an apple If the past tense in the complement clause Past2 functions like its main clause counterpart Pastl it should order the complement clause Reference Time RT 2 after TT2 ETZ Since sentence 3 must be understood to unambiguously locate ETl after ET2 RT2 must denote the same time as ET 1 since Past2 orders it after TT2 We can capture this by assuming 4 4 The RT of a complement clause is controlled bound by the main clause ETl The temporal interpretation of 3 is represented schematically in 5 5 UTRTl 7 TTlET 1RT2 7 TT2ET2 pastl past2 Thus 3 reports a prior event of Bill uttering la where the complement clause thatMax are an apple faithfully conveys the content of la Max are an apple 2 Finite SOT simultaneous interpretation ofpastin nite complement clauses A second difference between main clause and complement clause tense interpretation is illustrated by sentences 6ab 6 a Bill said that Max had eaten an apple b Bill said that Max was eating an apple If the complement clause past tense Past2 in 6ab were semantically equivalent to its counterpart in 3 we would expect that ET1 should control RT2 and that Past2 should order ETlRT2 after TT2 In 6a the complement clause perfect haveen should then locate TT2 after ET2 while the complement clause progressive being should locate TT2 within ET2 Thus we should expect temporal interpretations along the lines of 7ab with both sentences making reference to four distinct times 7 a UTRTl 7 TT 1ET 1RT2 7 TT2 7 ET2 6a pastl past2 perf b UTRTl 7 TT lETlRT2 7 TT2 6a pastl past2 l 7 7 7 prog X ET2 While it is possible to understand 6ab in this way in certain restricted circumstances discussed below the most salient interpretation of 6ab involves reference to only three distinct times where TT2 coincides with TT lET lRT2 as in 8 8 a UTRTl 7 TTlETlRTZ TT2 7 ET2 6a pastl perf b UTRTl 7 TT lET lRT2 TT2 6a pastl l lt ng X ET2 The interpretation associated with 8ab is one in which the pastshifting semantics of Past2 is entirely absent as though the complement clause past were being interpreted as a present zero tense rather than as a past tense For concreteness I assume that present tense expresses coincidence or simultaneity between RT and TT Thus 6a can be used to report a prior event of Bill uttering 9a and 6b can be used to report a prior event of Bill uttering 9b 9 a Max has eaten an apple b Max is eating an apple In the same way 10a can be used to report a prior event of Bill uttering either 10b or 10c 10 a Bill said that Max was in Paris b Max is in Paris c Max was in Paris The simultaneous interpretation of subordinate clause past is an instance of the phenomenon of SequenceofTense SOT involving distinctive correspondences between meaning and morphosyntactic form in complement clause tenses falling under the scope of a higher past tense especially in contexts involving indirect discourse reporting speech or mental attitudes There are two wellknown distributional restrictions on the simultaneous interpretation of the preterit past The rst is that it is possible only when the clause containing the past is embedded within a main clause containing another pastshifting past tense as in 6 The second is that it is possible only when the TT argument of past contains a stative predicate as in 10a or perfect or progressive aspect as in 6ab or a temporally quanti ed or habitual predicate as in 11 11 a Bill said that Max ate an apple every day b Bill said that Max ate apples Suppose that there is an aspectual supercategory STATIVE comprising conventional stative predicates predicates headed by perfect or progressive aspect and temporally quanti ed or habitual predicates This class contrasts with episodic eventive predicates of all aspectual subclasses including activities We need not be concerned here with the semantic principles de ning this grouping here Now the constraint on the simultaneous interpretation can be expressed descriptively as in 12a or 12b 12 a The internal argument of a temporalordering predicate expressing simultaneity must be the time of a STATIVE eventuality b The internal argument of a temporalordering predicate expressing simultaneity may not be the time of a nonSTATIVE episodic eventive eventuality The STATIVE constraint in 12 applies to other simultaneous tense interpretations as well In particular it also applies to uses of nite present in sentences such as l3ab as is well known 13 a Max eats an apple b Bill will say that Max eats an apple c Max is eating an apple d Bill will say that Max is eating an apple e Bill will say that Max eats apples every day Here the TT must contain a STATIVE predicate as in l3ce Sentences l3ab are anomalous they cannot be interpreted as nonprogressive analogues of l3cd with the present tense conveying simultaneity between RT UT and TT ET in 13a and between RT2 ET 1 and TT2 ET2 in 13b Sentences like 13a are appropriate as headlines where the understood tense is that of a recent past pastshifting rather than simultaneous or as captions on photographs or illustrations where there is no interpretation of simultaneity between the time of the event depicted and any other time I have suggested elsewhere Stowell 1995a 1995b 2006 that the pastshifting and simultaneous interpretations of the English preterit past in sentences like 6ab and 10a involve two distinct tenses the pastshifting reading involves a true past tense while the simultaneous reading involves a distinct zero or nonindexical presen tense I call the former tense PAST and the latter tense PRESENT While this might suggest that the finite preterit past is ambiguous between two lexical meanings PAST and PRESENT I suggest instead that past should be thought of as a temporal analogue of a determiner heading a time or event denoting expression but also incorporating a polarity marker indicating that the phrase it heads falls within the semantic scope of a true PAST tense On this view true semantic pastshifting PAST tense is covert null past is just the head of the TT argument falling under the scope of PAST When past occurs as the head of a main clause TT the true tense of the clause must be PAST in order to license past But when past occurs as the head of a complement clause TT it can be licensed by falling under the scope of a main clause PAST in which case the covert tense of the complement clause is free to be either PAST pastshifting or PRESENT simultaneous Completing the picture the morpheme present does not convey PRESENT simultaneity rather it is also the head of a TT argument but it conveys the opposite polarity relation of past the TT that it heads may not fall within the semantic scope of PAST When a complement clause contains past and the TT is STATIVE as in 6ab and 10a the tense is normally interpreted as if it were simultaneous PRESENT as we have seen But a complement clause containing past and a STATIVE TT can also be understood to contain a true pastshifting nonSOT PAST locating the matrix event time after the TT thus resembling the interpretation of past with an episodic eventive TT in 3 As noted by Boogaart 1995 however this is possible only when the subordinate clause TT is understood to be simultaneous to a time already under discussion in the prior discourse I will refer to this discoursesupplied time as the DT For example 10a repeated here can be used to report Bill s testimony in a criminal trial 10 a Bill said that Max was in Paris In this context the DT is the time of the alleged crime and Bill asserts that Max was in Paris at the DT Thus even on the pastshifted reading of past the STATIVE TT must be understood to be simultaneous with some other time The same is true with all other subtypes of STATIVE predicates they must normally be construed as simultaneous with some other time either with the main clause ETl when the tense conveys simultaneity or with a DT This raises the question whether the morpheme past ever conveys actual temporal pastshifting with TTs of STATIVE eventualities as it clearly does with TTs of episodic eventive eventualities The answer is yes On the pastshifted readings of sentences like 6ab and 10a where the TT is simultaneous with a DT the DT must be understood to be prior to the main clause ETithe time at which Bill spoke in 10a This must be the effect of the pastshifting interpretation of past within the complement clause since without this it should be possible for the TT to be simultaneous with a DT that is subsequent to the matrix ET It is a matter of controversy exactly how SOT works and whether the phenomenon is restricted to intensional contexts associated with predicates of speech belief and modality In this paper I will ignore these issues as much as possible focusing instead on the parallel between SOT interpretations of past in nite complement clauses exemplified in 6ab and cases involving non nite haveen to which I now turn 3 In nitival tense In nitival clauses are traditionally assumed to differ from nite clauses in three major respects First in nitival clauses at least in English lack any overt manifestation of subject verb agreement Second in nitival clauses at least in English lack overt nominative subject DPs the subject DP is typically either absent or null though in some cases nonnominative subjects are possible Third infinitival clauses are widely though not universally believed to lack tense That infinitives do not contain tense is of course the traditional view It directly accounts for the obvious fact that conventional tense af xes do not appear in them On the other hand from a semantic point of view in nitives can express the same basic temporal ordering relations that are conventionally expressed in simple nite clauses by past present and the future modal will To convey pastshifting infinitives use the bare perfect composed of the root form of the auxiliary have and a past participial complement haveen as in 14a To convey simultaneity 14b or futureshifting 14c in nitives require no additional morphology at all 14 a Sam believed Mary to have left Sam believed Mary to be in Paris c Sam expected Mary to leave So are in nitives really tenseless or do they harbor tenses after all More concretely do in nitives contain overt or covert syntactic elements that should be categorized as tenses conveying the semantic notions of past present and future The answer depends rst and foremost on how one defines tense From a semantic perspective three main ideas have been advanced 15 i Tenses express temporal shifting or lack thereof with respect to a Reference Time RT the RT is the Utterance Time UT at least in main clauses ii Tenses are referential expressions analogous to pronouns but referring to times rather than to individuals iii Tenses should be distinguished from aspects in being absolute or indexical always taking the UT as the RT 15i expresses the traditional intuitive View that tenses express the basic notions of past present and future 15ii captures a number of syntactic parallels between tenses and pronouns and has been widely in uential especially in the semantics literature Advocates of this view generally also accept 15i but assume that the temporalshifting function is subordinate to the referential function taking temporal shifting to be the main component of a restriction on the reference of the tense 15iii is controversial both with respect to the analysis of English tenses and with respect to the analysis of socalled relative tenses in many languages These ideas have been adopted in various combinations in specific theories of tense that have been advanced in the literature To convince all tense theorists that infinitives contain tenses it would be necessary to show that in nitives contain elements conveying a semantics that corresponds to all three of 15iiii I will not do that in this paper butI do intend to argue that in nitival perfect behaves like a past tense by either of the first two de nitions 15iii The in nitival perfect plainly does not behave in a way consistent with 15iii but I do not believe that 15iii should be taken to be a necessary property of true tenses In what follows I will assume the theory of tense that I have advocated elsewhere Stowell 1995ab to appear The theory assumes that the functions associated with 15i and 15ii are syntactically dissociated from each other The temporal ordering function 15i is assigned to the category Tense T while the temporal reference function 15ii is assigned to the timedenoting arguments of tense notably to the TT argument Contrary to 15iii I assume that tenses are not intrinsically absolute or indexical indexical interpretations of subordinate clause tenses arise when the application of overt or covert LF movement applies moving the subordinate TP or CF out of the scope domain of one or more higher tenses This type of movement can be triggered by various factors including the need to express de re reference To account for the phenomenon of SOT I further assume 16i iii 16 i The English tenses PAST and PRESENT are covert phonetically null ii The morphemes past and present are temporal analogues of determiners the heads of TT arguments referring to times iii past and present differ from each other in encoding a scopal polarity relation to PAST past must fall under the scope of PAST whereas present may not By 16i ii the English finite tense morphemes past and present are not true tenses expressing temporal ordering relations but rather heads of the TT arguments of tenses The pastpresent contrast expresses a scope relation to true PAST tense similar to the traditional account of the anysome contrast that it expresses a scope relation to negation or a downwardentailing operator This approach agrees with Partee 1973 and Enc 1986 1987 in claiming that the English morphemes past and present are the heads of timedenoting expressions as in 15ii but it disagrees with their accounts in claiming that these morphemes do not directly express any temporal ordering function 15i even as a restriction on the referential function It is sometimes asserted that pastshifting tenses have an absolute indexical tense interpretation whereas pastshifting aspects have only a relative timeshifting interpretation The absoluterelative distinction hinges on whether the tense is interpreted indexically or not in our terms this depends on whether the RT of the tense denotes the actual UT or some other time such as the main clause ET But the finite preterit past in the subordinate clauses in 3 repeated here has a relative rather than an absolute interpretation so it cannot be that only absolute tenses are true tenses 3 Bill said that Max ate an apple The same is true of past and present in examples like l7 17 a John willmight say that Max tricked him b John willmight say that he is thirsty Furthermore present and past tenses in Japanese have relative rather than absolute interpretations when they occur in subordinate clauses as in lSab just like haveen in an English infinitival complement The same is also true of tenses in relative clauses in Japanese as in 18c 18 a Taroowa cpHanakoga Tookyooni ita to itta TaroTOP HanakoNOM TokyoLOC bePST COMP sayPST Taro said that Hanako was had been in Tokyo Past shifted only b Taroowa Hanakoga Tookyooni iru to itta TaroTOP HanakoNOM TokyoLOC beNONPST COMP sayPST Taro said that Hanako was lit is in Tokyo Simultaneous c Taroowa waratte iru otokoo mita TaroTOPIC laughing beNONPST manACC seePST Taro saw a man who wasis laughing The same is true of tenses in many languages One can of course maintain that none of these are true tenses because of their nonindexical character but given the nonindexical character of the English tenses in 3 and 17 it is doubtful that the indexicality criterion 15iii can be maintained Tenses in relative clauses have been claimed to have an indexical interpretation but Abusch 1988 showed that this was not the case for relative clauses construed de dicta or de se suggesting that the scope construal of the relative clause is responsible for the indexical tense interpretation when the relative clause is construed de re I will therefore assume that tenses do not have to be indexical in order to count as true tenses In Section 4 I will argue that infinitives containing the bare perfect must be assumed in some cases to contain a covert counterpart to the same semantic formative PAST that is associated with nite clauses containing the preterit past If true tenses are temporal ordering predicates as in 15i then it is the presence or absence of these elements rather than the morphemes past and present that determines whether in nitives are tensed or tenseless Those who would defend a referential semantics for tenses as in 15ii might object that this criterion alone is insufficient Although I have suggested that the temporal reference function should be associated with the TT argument of a tense rather than with the tense itself I will argue in Section 5 that the in nitival perfect behaves like the finite preterit past in functioning as the head of a referential TT argument The argument is based on the observation that the in nitival perfect behaves like the nite preterit past in exhibiting a simultaneous SOT interpretation The broader conclusion is that infinitival clauses containing the perfect must be assumed in some cases at least to contain a past tense regardless of whether one assumes a predicative theory of tense of the sort I have advocated or a referential theory of tense of the ParteeEnc variety 4 The ambiguity ofthe infinitival perfect 41 Past tense vs perfect aspect If infinitives were really tenseless the nonfinite perfect in 19 would have to convey only an aspect and not a true PAST tense 19 Max believes Sam to have left However there is little empirical content to this claim unless one can show that the semantics associated with the nite preterit past morpheme is fundamentally different from that of the in nitival perfect in a way that follows naturally from the assumption that past conveys a true tense and that the in nitival perfect does not The idea that the infinitival perfect is or can be a true past tense was proposed by Hoffman 1966 Working within the framework of the Standard Aspects theory Hoffman assumed that infinitival clauses are derived transformationally from finite clause Deep Structure sources He derived infinitival haveen from three distinct nite sourcesithe preterit past 851 the present perfect hasen and the past perfect hadeno that 20 corresponds to any of 2lac 20 Caesar is believed to have lived in Rome 21 a It is believed that Caesar lived in Rome b It is believed that Caesar has lived in Rome c It is believed that Caesar had lived in Rome Deep Structure was assumed to be the sole syntactic level of representation at which semantic interpretation occurs the syntactic locus of the syntaxsemantics interface in Minimalist terms Hoffman s theory thus predicted that the in nitival perfect should be able to behave like any of the nite tense constructions in 2lac with respect to semantic tests distinguishing the tense constructions from each other It also predicted that the in nitival perfect should display ambiguous syntactic behavior corresponding to its three sources depending on the stage in the derivation at which the in nitive is created and distinctions among the three distinct tense constructions are neutralized Before proceeding further it should be noted that there are two very different ways of interpreting Hoffman s insight theoretically First it could be as he has it that the in nitival perfect has three distinct subtypes corresponding more or less directly to its three finite counterparts in 21 Translating his claim into the present framework this would imply that in nitival perfect clauses may contain any of the following three combinations a a past tense b a present tense and a perfect c a past tense and a perfect All three interpretations involve the presence of a tense within the in nitive a past tense in a and c and a present tense in b This interpretation of the ambiguity of 20 assumes of course that in nitival clauses may contain tense An alternative interpretation of 20 and 21 compatible with the traditional assumption that in nitives are tenseless would be that the infinitival perfect in 20 is simply vague rather than ambiguous along the dimension of the distinctions among the finite tense constructions in 21 Of course some combination of the two approaches might turn out to be correct 42 Speci c past time adverbs infinitival perfect as past Hoffman s diagnostic tests were stated as descriptive generalizations and were not given an explicit syntactic or semantic analysis First to show that the infinitival perfect can behave like the preterit past and unlike the present perfect he pointed out that it can cooccur with a time adverb that designates a past time point e g at 3 pm yesterday This is the familiar restriction on the English present perfect involving referential definite pasttime adverbs illustrated in 22 22 a He came last Tuesday b He has come last Tuesday c He is rumored to have come last Tuesday It should be noted that the past perfect and future perfect both differ from the present perfect in not being subject to this restriction Still the in nitival perfect in 22c cannot naturally be interpreted as either a past or future perfect so Hoffman is probably correct in claiming that the in nitival perfect in 22c corresponds semantically most closely to the preterit past But whether the test in 22 speci cally diagnoses the presence of a syntactic or semantic counterpart to a past tense in the infinitive is another matter Insofar as the de nite time adverb restriction applies specifically to the present perfect it could be that the in nitive simply contains a bare tenseless perfect which might be expected to behave like the nite future and past perfects in not being subject to a restriction that applies only when the present tense is involved Thus the test in 22 turns out not to be decisive 43 Now in nitival perfect as present perfect Hoffman s second test was intended to show that the in nitival perfect can behave a finite present perfectiand unlike a nite preterit past or past perfectiin being compatible with the timeadverb now Actually I think that Hoffman s description of the facts is insuf ciently fine grained The past perfect is compatible with a nonindexical relative narrative pasttime interpretation of now indicated by A in 24c and 25c the same interpretation arises somewhat marginally without a prior discourse context with the preterit past in 25a and with the future perfect in 24d and 25d 23 a He is reportedbelieved to have drunk a gallon of vodka by now b He is allegedbelieved to have nished eating now 24 a It is reported that he drank a gallon of vodka by now b It is reported that he has drunk a gallon of vodka by now c AIt is reported that he had drunk a gallon of vodka by now d AIt is expected that he will have drunk a gallon of vodka by now 25 a AIt is alleged that he nished eating now b It is alleged that he has finished eating now c AIt is alleged that he had nished eating now d AIs is expected that he will have nished eating now Example 24a is worse than 25a presumably because the preterit past in combination with an episodic eventive VP is incompatible with any completive adjunct PP headed by by as illustrated by 26 26 It is reported that he drank a gallon of vodka by 5 o clock In any case the correct descriptive generalization about now seems to be that an indexical interpretation is possible if the clause containing it contains the present perfect but not if it contains the preterit past or the past or future perfect Does this tell us that the in nitival perfect in 23 contains a counterpart to the present perfect Unfortunately the answer is less clear than what Hoffman claimed If the test speci cally diagnoses the presence of a present tense within the clause then the answer is yes If on the other hand the test simply diagnoses the absence of a past or futureshifting tense ie the absence ofa tense or modal shifting the topic time TT away from the present UT then the answer is no since the hypothesis that infinitives are tenseless is compatible with the latter claim Can one choose between these views To resolve this the rst step is to formulate a more articulate theory of the basis of the relevant constraint on the indexical interpretation of now When a time adverbial occurs with a perfect it can in principle associate either with the ET or with the TT the socalled result state time In the case of an indexical adverb like now the only option is the result time When the TT is the complement of a past tense on either a pastshifted or simultaneouspast reading or of a future modal the TT cannot refer to the UT and indexical now is excluded The question of whether the in nitival perfect in 18 contains a counterpart to the present tense thus hinges on whether a present tense is required within the clause in order for perfect to provide a TT that the indexical now can associate with I see no reason to believe that this must be the case so it must be concluded that Hoffman s second test is also indecisive 44 Double past time adjuncts In nitival perfect as past perfect Hoffman s nal test intended to show that the infinitival perfect can correspond uniquely to the finite past perfect with a particular combination of temporal adjuncts as in 27 and 28 27 He is rumored to have seen her only once before when I met him 28 a It is rumored that he saw her only once before when I met him b It is rumored that he has seen her only once before when I met him c It is rumored that he had seen her only once before when I met him The test works because the two adjuncts have to associate with distinct past times The adjunct whenclause in 27 and 28c is associated with the TT the perfect result time It contains a past tense so the TT must itself be in the past This is what excludes the present perfect in 28b The first adjunct only once before is existentially quanti ed binding an event or ET variable and it intemally locates ET in the past relative to another time T ET is before T Now T is covert but it is anaphorically bound by the TT the perfect result time in 27 and 28c Since the two adjuncts modify distinct past times they are incompatible with the preterit past in 28a since it lacks an aspectual auxiliary its ET functions as its TT and it fails to provide two distinct timedenoting arguments for the adjuncts to modify Now since the in nitival perfect in 27 is compatible with this combination of adjuncts it must provide two distinct timedenoting arguments for the adjuncts to modify In principle the pastshifting perfect should provide them Must we assume that the in nitive contains a past tense in addition to the perfect At first glance the answer might appear to be no on the following grounds as long as there is no present tense in the in nitive one might assume that the TT the perfect result time is free to refer to any time past or present if it picks out a past time it can be compatibly modi ed by the pasttense whenclause and its ET can of course be bound by the existentially quantified adjunct On closer inspection however the answer must be yes since an ECM infinitive with a STATIVE TT cannot normally receive a pastshifting interpretation 29 He is rumored to be tall Here the content of the in nitival clause must have an indexical presenttense interpretation Now while one can attribute a pastshifted ET to the perfect aspect in the infinitive in 27 once cannot attribute a pastshifted TT to it on the basis of the perfect aspect alone Therefore the in nitival perfect in 27 must contain a pastshifting tense to locate the TT in the past in 27 exactly as Hoffman claimed The same is true of somewhat simpler examples like 30 30 John is believed to have already left when I met him Thus it seems that Hoffman s third test provides positive evidence in favor of the view that in nitival perfects must be capable of encoding a past tense at least in combination with a pure perfect aspect How exactly the in nitival perfect manages to encode two pastshifting tenses in 27 and 30 will be addressed in Section 6 Chris Collins personal communication has suggested that the examples in 27 and 29 might be assumed to involve simple iteration of the perfect as in 3 l with subsequent reduction of have had to have in 27 and 29 as a type of haplology 31 a He is rumored to have had seen her only once before whenI met him b John is believed to have had already left when I met him Collins points out correctly that examples like 31 are common in many dialects and are abundantly provided by web searches in examples such as the following 32 Unfortunately a company appears to have had already gone out of business by t en Though many speakers including me nd examples like 31 and 32 utterly ungrammatical they are clearly attested for many speakers I do not know whether this correlates with geographically de ned dialects or is a matter of idiolectal variation Even granting a haplology analysis of 27 and 29 along the lines that Collins suggests one might still conjecture that true iteration of the perfect is not in general permitted eg in nite clauses in dialects that allow 3 l and 32 in which case the rst perfect in these examples might be argued to correspond to an independent past tense However examples of iterated perfects in finite clauses turn up with surprising frequency in web searches with both present perfect and past perfect in examples like those in 33 33 a Hoboken has had begun planning discussions about options for clearly identifying certain routes as throughtraf c bypasses b The HSC had approved the funding to begin in August although Liverpool had had begun work already So it seems that iteration of the perfect is possible in some dialects and that this is a plausible analysis of the in nitival examples in 3 l and 32 But whether 27 and 29 in the standard dialect involve an iterated perfect that undergoes haplology is another matter The haplology rule in question would have to be arbitrarily con ned to in nitives since the present perfect in finite clauses can never be construed as an iterated perfect in examples like 34a 34 John has left not It has been the case that John had left It is less easy to rule out an iterated perfect interpretation for the past perfect in a finite clause but if it were possible then examples such as 35 ought to be possible with the rst adjunct associating with the event time the time at which John saw her and the second and third adverbials associating with the resulttimes of the two perfects 35 John had seen her only once before when I met him when I left I nd such examples impossible to parse suggesting that the putative iterated perfect and its associated rule of haplology is disallowed in finite clauses While it is hypothetically possible that this is allowed speci cally in in nitives I consider this possibility unlikely 5 In nitival Sequence of Tense 51 Sequence of Perfect A different kind of argument for the presence of a past tense within infinitives comes surprisingly and ironically from cases where the in nitival perfect seems to lack any past shifting interpretation at all These are cases hitherto unnoticed to my knowledge where the in nitival perfect behaves like an SOT preterit past allowing a simultaneous relative present or Zero tense interpretation when embedded under a past tense main verb 36 a Caesar had actually believed his wife to have been in Rome at that time b Caesar had once alleged Pompey to have been a scoundrel c After the battle Caesar appeared to his soldiers to have been unwell Although a pastshifting interpretation for the infinitival perfect is possible here it is not required 36ab are ambiguous along the simultaneous vs pastshifted interpretation in exactly the same way as their nite counterparts in 37ab are 37 a Caesar had actually believed that his wife was in Rome at that time b Caesar had once alleged that Pompey was a scoundrel c After the battle it appeared to his soldiers that Caesar was unwell Although the pastshifted interpretation is favored in 36 as the simultaneous interpretation is favored in 37 both interpretations are possible in both cases I find that the simultaneous reading is slightly more natural in 36ab when the main clause contains the past perfect though the reading is still possible with the simple past Comparing 36ac to 38ac we find that the examples in 38 are temporally unambiguous having only a simultaneous interpretation 38 a Caesar had actually believed his wife to be in Rome at that time b Caesar had once alleged Pompey to be a scoundrel c After the battle Caesar appeared to his soldiers to be unwell Many speakers prefer 38 over 36 to express a simultaneous reading but 36 allows it too Given the analysis of SOT summarized above the facts suggest that the in nitival perfect like the nite preterit past actually functions not as a pastshifting tens e but rather as the referential head of a TT argument incorporating a PAST polarity marker indicating that it falls under the scope of a covert PAST On the pastshifted reading the FAST licensing the in nitival perfect polarity item resides within the infinitive on the simultaneous reading the in nitival perfect polarity item is licensed by the main clause PAST and the infinitive contains a covert PRESENT tense instead There does seem to be a very subtle difference between 38 and the simultaneous reading of 36 though it is not clear to me precisely what is involved This difference may be related to the contrast between 39 and 40 39 a John told me yesterday that next week his mother would believe him to have been sick b John told me yesterday that next week he would claim to have been sick 40 a John told me yesterday that next week his mother would believe that he was sick b John told me yesterday that next week he would claim that he was sick Whereas 40 allows a simultaneous present tense interpretation of the most deeply embedded clause relative to the event time of the clause immediately containing it this does not seem to be possible for the in nitival perfect in 39 Examples like 40 originally due to Kamp and Rohrer 1983 were cited by Abusch 1988 as evidence against the view that the simultaneous SOT reading of past actually involves an indexical past tense That an analogous simultaneous reading is apparently impossible in 39 might be taken as evidence for the opposite view This however would be a surprising conclusion to draw about the in nitival perfect since it would entail that the infinitival clause contains an indexical past tense where the finite clause in 40 does not In any case the available interpretation in 39 is not an indexical past but rather a pastshifted reading relative to the event time of the clause immediately containing it The same is true of 41 without the indexical adverb in the intermediate clause 41 Caesar told Mark Anthony that his wife would believe him to have been in Rome Rather it seems that the in nitival perfect disallows a simultaneous SOTtype reading when the futureshifting wall intervenes between the perfect and the FAST tense that licenses the SOT effect unlike the situation with finite past in 40 This suggests that there is a locality condition governing the licensing of infinitival SOT that does not constrain finite SOT In nitival control clauses often have a futureshifting tense interpretation relative to the event time of the main clause control verb In these contexts Stowell 1982 assumed the presence of a futureshifted tense within the in nitive Wurmbrand 2005 suggests that in such cases there is a covert infinitival counterpart to wall which she takes not to be a true tense These views turn out to be indistinguishable given a 43 Caesar had expectedhopedwantedpromised to be in Rome when his wife arrived In these contexts I find that a simultaneous nonpastshifted reading of the infinitival perfect is possible though it involves simultaneity with the futureshifted time introduced by the in nitival counterpart of wall rather than with the matrix event time associated with the intensional verb 44 Caesar had expectedhopedwantedpromised to have been in Rome when his wife arrived Example 44 has a counterfactual avor that is somewhat less favored in 43 so it is possible that when the in nitival perfect licenses a simultaneous reading in futureshifted in nitives the perfect is used to encode counterfactuality In this respect the infinitival perfect again resembles the preterit past in a nite clause which allows a presenttense construal in weakly counterfactual conditionals such as 45 45 If John was here he would be hiding somewhere Interpretations essentially parallel to 44 are also observed in finite complements as in 46a where the covert futureshifter is replaced by would formed by combining past with wall 46 a Caesar had expectedhopedpromised that he would have been in Rome when his wife arrived b Caesar had expectedhopedpromised that he would be in Rome when his wife arrived Thus it seems that the intervention effect that blocks the simultaneous reading of the in nitival perfect in 39 and 41 does not arise when the intervening futureshifter occurs in the same clause as the infinitival perfect I will leave it to future research to determine the nature of the intervention effect and whether its mitigation in 44 and 46a is due to the in nitival perfect being licensed by a counterfactual operator or a covert subjunctive mood within the in nitive Either way these simultaneous andor counterfactual interpretations associated with the in nitival perfect are parallel to the behavior of the preterit past in nite clauses and unlike the semantics usually associated with perfect aspect This is also consistent with the view expressed in lSii that the timedenoting aspect of past tense rather than its past shifting semantics is essential to its status as a tense 52 Sequence of tense triggered by the in nitival perfect Another way of using SOT to diagnose the status of the in nitival perfect as a variant of PAST tense is to show that it triggers SOT in finite clauses falling within its scope domain Brugger and d Angelo 1994 use this test to argue that the nite present perfect in Italian is ambiguous between two distinct interpretations they treat one of these as a true past tense and the other as a nonpasttense perfect aspect The latter they take to involve an abstract formative TERM indicating something like perfectiveness Only the former past tense usage of the present perfect licenses a simultaneous interpretation of a past imperfect tense in a subordinate clausal complement They cite a couple of other diagnostic tests distinguishing between the two interpretations of the perfect which correlate reliably with the SOTtriggering test Rather than citing their Italian examples and summarizing the somewhat complex interactions that they involve I will simply construct contrasting examples in English that seem to behave similarly In 47a the English perfect conveys a true pastshifting interpretation and licenses finite SOT in its complement in 47b on the other hand the English perfect apparently does not behave like a past tense with respect to SOT licensing the complement clause in 47b unlike its counterpart in 47a does not allow a simultaneous SOT reading 47 a John has often believedthoughtsaid that he was unhappy b John has now realizedaccepted that he was unhappy c John had already realizedaccepted that he was unhappy Applying this test to the infinitival perfect we see that it can behave like the preterit past and the pastshifting perfect in licensing SOT in a finite complement clause 48 a John is believedknownalleged to have claimed that he was unhappy b John is believed to claim that he was unhappy Once again the infinitival perfect behaves like the in nitive contains a past tense capable of triggering SOT ie of licensing a past polarity item in a finite clausal complement though in this respect it does not differ from the pastshifting reading of the present perfect in 47a 6 Tying up a loose end It remains to provide an account of the pastperfectlike interpretation of the in nitival perfect in examples like 27 and 30 repeated here 27 He is rumored to have seen her only once before when I met him 30 John is believed to have already left when I met him The most natural move to make is to assume that the infinitival perfect polarity item is licensed by two pastshifting tenses within the in nitive One way of thinking about this is to assume that one of these is a counterpart to a nite PAST while the other is a counterpart to the pastshifting semantics normally associated with the perfect Though on this view these pastshifting formatives are covert their presence does not come for free perhaps because of economy considerations their presence must be licensed by an overt PAST polarity element past in a nite clause and perfect in an infinitive This in turn raises the question whether all instances of the perfect in nite and non finite clauses alike functions as a PAST polarity item excluding cases where the perfect polarity item is licensed by TERM or by a subjunctivecounterfactual operator This does seem to be a viable option though it must be noted that when the perfect occurs in its finite form it never triggers a pastperfect interpretation analogous to what we nd in 27 and 30 But this can be accounted for by the fact that nite clauses must contain either past or present which either licenses a past tense independently past or excludes it present These polarity items are absent from in nitives so the non nite perfect is free to license two pastshifting tenses in examples like 27 and 30 Even in infinitives this interpretation is accessible only when two temporal adjuncts are present suggesting that economy considerations prevent the perfect polarity item from licensing more than one covert pastshifting tense unless the absence of a second pastshifting tense would cause the derivation to crash This implies that the pastshifting semantics associated with perfect aspect is really parallel to the pastshifting semantics associated with past tense since both can be 7 Conclusion In this article I have argued in the spirit of Hoffman 1966 that the in nitival perfect may function like the nite preterit past The particular instantiation of this idea adopted here is that the in nitival perfect like the preterit past is actually not a tense but rather a PAST polarity item serving as the head of a timedenoting expression rather than as a true past shifting tense The in nitival perfect can be licensed by a covert PAST residing inside or outside the infinitive The upshot is that in nitival clauses must be assumed to contain at least one type of tense namely PAST lending some support to the view that infinitives may contain other tenses as well simultaneouspresent or futureshifting In nitival clauses differ from nite clauses however in lacking an overt counterpart to finite present which as I have suggested encodes the opposite polarity relation to that expressed by past The latter element rather than PRESENT tense per se sometimes gives rise indirectly to indexical tense interpretations when it occurs under the syntactic ccommand domain of a higher PAST tense References Abusch D 1988 Sequence of tense intensionality and scope WCCFL 7 114 Boogaart Ronny 1995 Towards a Theory of Discourse Aspectuality Pier Marco Bertinetto Valentina Bianchi James Higginbotham and Mario Squartini eds Temporal Reference Aspect analActionality vol 1 Semantic anal Syntactic Perspectives Torino Rosenberg and Sellier 221236 Brugger G and M D Angelo 1994 quotAspect and Tense at LF The Italian Present Perfectquot Ms University of ViennaUniversity of Venice Enc Murvet 1986 Towards a Referential Analysis of Temporal Expressions Linguistics and Philosophy 94 405426 Enc M 1987 Anchoring conditions for tense Linguistic Inquiry 18 633657 Hoffman T Ronald 1966 quotPast Tense Replacement and the Modal Systemquot A Oettinger edMathematical Linguistics and Automatic Translation Cambridge Massachusetts Harvard University Harvard Computational Laboratory VII121 Reprinted in McCawley 1976 Notes from the Linguistic Underground Syntax and Semantics 7 New York Academic Press 85100 Kamp Hans and Christian Rohrer 1983 Tense in Texts39 in R Bauerle et al eds Meaning Use and Interpretation of Language Walter de Gruyter Berlin 250269 Klein W Time in Language London and New York Routledge 1994 Nakamura A 1995 On some temporal constructions in Japanese and the theory of tense Manuscript University of California Los Angeles Partee B Some Structural Analogies Between Tenses and Pronouns in English The Journal ofPhilosophy 70 1973 Prior A 1967 Past Present and Future London Oxford University Press Reichenbach Hans 1947 Elements of Symbolic Logic University of California Press Berkeley CA Book Reviews DOMINIC W MASSARO EDITOR University of California Santa Cruz TipoftheTongue Phenomena Gold Mine or Can of Worms TipoftheTongue States Phenomenology Mechanism and Lexical Retrieval By Bennett L Schwartz Mahwah NJ Erlbaum 2002 x 181 pp Cloth 3995 Tipof thetongue TOT phenomena have proven a gold mine for theories of memory retrieval and midlevel lexical and phonological language produc tion and have provided important insights into relationships between language and memory see especially MacKay amp Abrams 1996 TOT states also carry practical signi cance because they require theories of memory and language to address phenomena commonly observed in everyday life Speakersin the TOT state are temporarily unable to retrieve the full phonology for a word that they know and have successfully retrieved many times They often feel that they will soon recall the soughtfor word and can usually recognize it if presented to them They invariably retrieve the word s meaning and can often retrieve its syntactic category its stress pattern and number of syllables its initial sound or letter and its gender in languages such as Italian eg Brown amp McNeill 1966 Burke MacKay Worthley 8c Wade 1991 Miozzo amp Caramazza 1997 Vigliocco Antonini amp Garrett 1997 Alternate words that resemble the tar get in syntax meaning and phonology often come repeatedly and involuntarily to mind even though the speaker rejects these persistent alternates as inap propriate However persistent alternates decrease with aging even though young adults experience fewer TOTs than older adults and can accurately re port more phonological characteristics of the target word eg Burke et al 1991 Brown 8c Nix 1996 Heine Ober 8c Shenaut 1999 Maylor 1990 Laboratoryinduced TOTs have also provided a gold mine of theoretically important information For example recent production of a target word Rastle 8c Burke 1996 or aspects of its phonology in phonologically related words Games 8c Burke 2000 Meyer amp Bock 1992 reduces TOT likelihood and in creases resolution likelihOod for ongoing TOTs elicited in the laboratory However in Tip ofthe Tongue States Phenomenology Mechanism and Lexical Retriev al Schwartz describes his own laboratorybased results as mixed a gold mine of fascinating discoveries and a can of worms of perplexities and oddities p ix This review explores the origins of these perplexities and suggests direc tions for future research to separate the productive approaches from the not soproductive approaches described in this book 292 AMERICAN fOURNAL OF PSYCHOLOGY SUMMER 2003 Origins of the perplexities De nitional and procedural issues Schwartz is studying a phenomenon that hequot calls subjective TOTs in con trast to the objective TOTs studied by other investigators Because all TOT experiences are subjective and all TOT data are objective we prefer the label SchwartzT073 STOTs to capture the conceptual difference betweenTOTs and STOTs For Schwartz STOTs are the feeling of being On the verge of being able 39 to recall the answer that you cannot recall now p 3972 whereas the vast major ity of TOT researchers stipulate an additional criterion that participants know and can recognize the target word This seemingly minor conceptual difference has triggered important differences in theexperimental pr0cedures for elicit ing TOTs versus STOTs We review these procedural differences in detail because they carry far reaching implications for future research on STOT and TOTs In TOT studies such as Burke et al 1991 pilot tests ensure that partici pants are likely to know the target words for eliciting TOTs and de nitions are constructed to unambiguously specify precisely those words For each de ni tion participants respond either know indicating they haveretrieved the phonology for a word corresponding to the de nition don t know indicat ing inability to retrieve a word with that de nition or TOT indicating they know a word matching that de nition but are temporarily unable to recall its full phonology Participants who respond know give their word which ends their trial if it matches the target word Ifit does not they receive a fourchoice recognition test to determine whether they can discriminate the target word from semantically similar neighbors of the same word class Participants who respond TOT receive the same recognition test with none of the above as an additional response alternative for each item Statistical analyses focus mainly on TOT responses where the target word is accurately identi ed Now consider the quite different procedures for eliciting and analyzing STOTs One is the illusory STOT paradigm Schwartz 1998 see also Schwartz Travis Castro 8c Smith 2000 Participants receive 100 questions with three 7 possible responses an answer TOT or don t know Some 20 of the questions are unanswerable eg What is the capital city of BormeaP and if they answer an unanswerable question Clearly indicating that they are not in a TOT state they are told that their answer was incorrect They are then asked Are you in the TOT state This demand characteristic violates the standard conception that TOTs 39occur when speakers fail to retrieve a speci c word that they believe is appropriate Schwartz provides no j usti cation for this highly unusual procedure After all 100 questions these STOT procedures are repeated for all 100 questions and participants who respond don t know or answer incorrectly indicate again whether they are having a TOT and rate their feelings of imminent recall andtheir imagined ability to select the correct answer from among seven or more alternative answers none of which are cor rect for unanswerable questions Finally for answerable questions only par ticipants choose one of the recognition alternatives 39 Another novel STOT paradigm involves new learning Schwartz 8c Smith 1997 Schwartz 1998 Participants rst learn an imaginary name for an imag inary animal with some property that later serves as a retrieval cue for recall BOOK REVIEWS 293 ing the animal name Participants who respond can t recall are asked wheth er they are experiencing strong feelings regarding eventual recall and wheth er they can recall the rst letter of the animal name or any related informa tion from the study phase In the subsequent recognition test all participants regardless of prior recall performance are instructed to choose39the appro priate animal name among three foils other names from the study session Unsuccessfully recognized targets are included in STOT analyses The issue raised by the differing procedures for eliciting TOTS versus STO Ts is this Do identical processes underlie the retrieVal of familiar words the theoretical domain of TOTs and the learning of pseudowords one theoret ical domain for STOTs Schwartz provides no reason for assuming that STO Ts and TOTs tap identical processes and we will cite data suggesting that they do not Until this issue is resolved it seems prudent to distinguish TOTs from STOTs on empirical theoretical and practical grounds TOTs are objects of practical relevance because they involve familiar words from everyday life whereas STOTs involving unknowable words and neverencountered pseudo words have unknown practical signi cance It may not matter that Schwartz characterizes STOTs as one of those illusive sic oddities of human cognition p ix but for the older adult experiencing worrisome increases in the frequen cy of selfproduced TOTS in everyday life see Burke et al 1991 it is impor tant not to characterize TOTs as either illusive or odd STOTs Some illustrations of the problems Schwartz is of course free to study any aspect of behavior using whatever eliciting conditions he chooses If Schwartz succeeds in demonstrating unam biguous causal regularities that further our understanding of mind we applaud him However artifact free and unambiguous relationships between STOTs and the mind remain to be demonstrated as the following problems illustrate Contradictory results STOTs and experienced emotions This book recognizes the importance of relationships between emotion and memory see also MacKay et al 2002 but STOT results to date are contra dictory In a diary study the rated degree of frustration that participants expe rienced during STOTs correlated negatively with the ability to resolve STOTs in everyday life p 39 whereas in a laboratory study experienced frustration during STOTs correlated poSitively with resolution and recognition p 39 In short the relationship between STOTs and emotion approximates the can of worms rather than the gold mine Demand characteristics A methodological issue When Widner Smith and Graziano 1996 misinformed participants that TOTinducing questions would be easy participants reported more TOTS than accurately informed participants but equivalent expectations of target recog nition or feeling of knowing FOK From this Schwartz concludes that demand characteristics cause TOTs and that TOTS are dissociable from memory retrieval processes However report of a TOT does not necessarily correspond to being in the TOT state Demand characteristics may alter TOT reports without caus 294 AMERICAN fOURNAL 0F PSYCHOLOGY SUMMER 2003 ing TOTs per se just as response bias can in uence memory measures with out in uencing memory per se Demand Characteristics are a methodological issue that any laboratory study must address Schwartz does not address this methodological issue in his own experiments and he is mistaken in suggest ing that it contributes to the primary goal of TOT research understanding the mechanisms underlying the everyday occurrence of TOTs Confounds and experimental artifacts Illusory STOTs Schwartz p 118 suggests that illusory TOTs ie STO Ts for unanswerable questions provide strong support for two propOsitions repeated throughout the book that people do not directly experience being in the TOT state but indirectly infer the existence of a word in memory using fallible clues and that subjective experienCes associated with STOTs can and should be studied separately from objective processes such as recognition of the target word and partial retrieval of its phonology However simpler and less problematic interpretations are possible One is that illusory STOTs represent incorrect TOTs involving a word other than the target a frequent occurrence in TOT studies eg 23 of all responses in Burke et al 1991 Incorrect TOTs have four basic causes misperception eg participants might misread Bormea in What is the capital of Bormea as Borneo or Burma misrepresentation eg participants might misrepre sent What is the name of the moon orbiting Mercury as What is the name of the moon orbiting MarSP see Shafto 8c MacKay 2000 for procedures ap propriate to studying this Moses phenomenon and imprecise and or incom pletely encoded questions eg participants who incompletely encode the de nition a circle or any indication of radiant light around the heads of divinities saints sovereigns in pictures medal etc might respOnd halo in correct rather than nimbus correct p 67 Three procedures can determine whether illusory STOTs are incorrect TOTs The rst entails reanalysis of the know responses in Schwartz et a1 Partici pants who responded jakarta or Bagan to What is the capital city of Bormea must have misperceived or misrepresented Bormea as Borneo or Burma The second involves postexperimental requests that participants ex plain their incorrect responses The third involves a multiple choice recog nition test for illusory STOT questions If participants misrecognize What is the capital of Bormea as What is the capital of Borneo or What is the capital of Burma their prior response is an incorrect TOT that carries none of the theoretical baggage of illusorySTOTs Schwartz et a1 adopted none of these procedures for ruling out the incorrect TOT hypothesis 39 Several additional factors render the incorrect TOT account of illusory STO Ts especially plausible One is the extraordinarily high rate of TOT responses and low rate of correct target recognition in STOT studies as compared to those in TOT studies For example the TOT response rate for answerable questions was 44 in Schwartz 1998 compared with the typical 99 in Burke et a1 1991 whereas the rate of correct target recognition was 40 in Schwartz and 77 in Burke et a1 These differences suggest that the in ated STOT rates in BOOK REVIEWS 295 Schwartz re ect the occurrence of incorrect STOTs and the same may be true for illusory STOTs involving unanswerable questions Consistent with this interpretation are the many other differences between illusory STOTs naturally occurring and experimentally induced TOTs and standard STOTs First TOTs in everyday life are usually resolved eg 98 resolution for the oldest adults in Heine et al 1999 see also Burke et al 1991 whereas illusory STOTs are unresolvable in principle and never occur in every day life Second Schwartz reported that rated strength emotionality and im minence differed reliably for STOTs to answerable versus unanswerable ques tions Third unlike young adults almost every older adult in a study of illusory STOTs detected the falsity of unanswerable questions without reporting STO Ts p 140 This indicates quota major difference between TOTs and illusory STO Ts because no published study has reported signi cantly more TOTs for young than older adults Finally Schwartz himself postulates a fundamental difference between TOTs and illusory STOTs by suggesting that the same factor the great er knowledge of older adults in uences these phenomena in opposite ways causing an agelinked decrease in illuSory STOTs p 140 and an age linked increase in genuine TOTs p 139 I The amountof information effect The amount ofinformation effect re ported in Schwartz and Smith 1997 represents a potentially important con tribution of the STOT newlearning paradigm During 10 3 and 5 8 study phas es three groups of participants learned to associate the name of an imaginary animal with a country plus varying amounts of additional information no ad ditional information for group 1 a picture of the imaginary animal for group 2 and its picture size and imagined diet for group 3 In the cuedrecall phase participants received the country names and either recalled the associated an imal names or reported feelings of imminent recall Results indicated that the amount of information associated with imaginary animals during study covar ied with STOTs without in uencing overall correct recall of the target animals These results allow two possible interpretations The simplest is that provid ing more information during learning increased the likelihood during recall that participants in a don t know state incorrectly responded STOT This false STOT hypothesis is readily checked by comparing across conditions false reports of STOTs ie cases in which participants chose an incorrect animal name on the recognitiOn test following attempted recall Unfortunately how ever Schwartz and Smith did not report recognition data for STOT responses by amount ofinformation condition Nevertheless data on partial phonologi cal recall that Schwartz and Smith did report is consistent with this falseSTOT hypothesis Recall of the rst letter of the imaginary animal names did not vary with amount of information This suggests that increased STOTs in the medi um and highinformation conditions involved don t know rather than gen uine TOT like responses because the rst letter for a target word is commonly reported during genuine TOTs but not following don t know reports see eg Burke et a1 1991 Moreover it makes good theoretical sense for don t know responses to increase with amount of information in this newlearning para digm since people surely learn two pieces of information better than ve pieces 296 AMERICAN jOURNAL 0F PSYCHOLOGY SUMMER 2003 of information during a xed study time Although correct recall of the target word did not differ across information condition recall of only the rst three letters counted as correct recall in Schwartz and Smith allowing the illusion of xed recall if participants reported more complete words in low than high information conditions 39 The second possible interpretation suggested by Schwartz is that this ef fect cannot be explained in theories such as the transmission de cit hypothe sis Burke et a1 1991 that provide a detailed account for much of the data on39TOTs schwartz p 64 This suggestion is accurate because current appli cations of the transmission de cit hypothesis involve TOTs for familiar words with established connections rather than processes for forming new connec tions to represent pseudow39ords However no explanation is needed in theo ries of TOTs if the amountofinformation effect is artifactual Moreover if STOTs and TOTs turn out to be fundamentally different phenomena explain ing them will require different theories and Schwartz in fact proposes a fun damentally different theory namely that STOTs are caused by any informa tion that is retrieved including information that may actually be tangential to the soughtafter target p 68 Two well established aging effects contradict this retrievedinformation theory Young adults report more partial informa tion and more persistent alternates than older adults and should therefore report more rather than fewer TOTs than older adults under this theory Gratuitous incoherent unparsimonious and counterfactual hypotheses Schwartz proposes three hypotheses that trivialize age linked TOT effects The rst concerns stress Schwartz suggests that age related increases in TOTs re ect the increased stress that older adults experience when their retrieval system does not appear to be functioning as ef ciently as when they were young er p 43 However this hypothesis is gratuitous without independent evi dence for agelinked increases in stress and for effects of stress on TOTs in coherent barring some account of how stress could selectively increase TOTs without reducing overall correct recall unparsimonious assuming a less ef cient retrieval system alone suf ces for explaining agelinked increases in TOTs39 see Burke et al 1991 and counterfactual ratings of worry and fa tigue at TOT onset were higher for middleaged than for older adults in Burke et a11991 The second hypothesis concerns social motivation According to Schwartz older adults report more TOTs than young adults because of self perceived declines in memory that suggest they should p 145 This social motivation hypothesis predicts more incorrect TOTs for older than young adults on post experimental tests of target recognition whereas rates of incorrect target rec ognition39following TOT responses are age constant see eg Burke et a1 1991 The third hypothesis concerns knowledge Schwartz p 139 suggests that TOTs increase with aging because of the greater knowledge of older adults One problem with this hypothesis is that older adults are more likely to experience TOTs for proper names than for abstract words eg adjectives verbs even though abstract words represent a much larger knowledge base than proper names Burke et a1 1991 Another problem is that TOTs increase with aging BOOK REVIEWS 39 297 when knowledge is equated for young and older adults ndings not reviewed in this book For example when participants named pictures of movie stars in Cross and Burke 2001 TOTs increased with aging despite equivalent knowl edge for young and older adults measured as the mean number of correct responses to the pictures TOTs for names of famous people also increased with aging in Burke et al 1991 even though familiarity with famous names was slightly greater for young than for older adults Theoretical problems The theoretical framework in this book is based on two misinterpretations of Tulving 1989 One is that SChwartz mistakes Tulving s focus on unconscious processing as relevant to the contents of consciousness eg FOK a very dif ferent issue Another mistaken claim attributed to Tulving p 16 is that sepa rate processes cause memory retrieval and content awareness Tulving wrote only that phenomena such as nonconscious learning surprise psychologists who generally View cognition behavior and awareness as highly correlated How ever correlation never implies causation These mistakes aside what is lacking in STOT research are theoretical ideas as to how subjective feelings such as imminent recall originate and how feelings that often accompany TOTs relate to memory retrieval processes Separating the gold mine from the can of worms Prospects for further research Schwartz treats STOTs and TOTs as indistinguishable phenomena developed under differing theoretical frameWorks However we have shown that STOTs differ empirically from the TOTs that occur in everyday life and further research is needed to test for differences between experimentally induced TOTs and STOTs as empirical phenomena For example reanalyses of current STOT data are needed to separate out the standard TOTs in these data with correct target recognition in the posttest only then can we compare the TOT results of Schwartz with those of other studies and assess what aspects of current STOT data re ect artifacts and unusual demand characteristics of the novel procedures for eliciting STOTs Further theoretical development is also needed to determine whether additional research using STOT procedures is worth doing Note Don MacKay gratefully acknowledges support from NIA grant ROlAG 09755 The authors thank Dr Deborah Burke for helpful comments on an earlier draft Jennifer K Taylor Psychology Department University of California Los Angeles Los Angeles CA 90095 1563 E mail jktayloruclaedu Donald G MacKay Psychology Department University of California Los Angeles Los Angeles CA 90095 1563 Pr esupposition 2week course The New YorkSt Petersburg Institute of Cognitive and Cultural Studies Philippe Schlenker UCLA amp Institut J eanNicod Pr esupposition Approximation A presupposition of S is a condition that must be met for S to be true or false Presuppositions a John knows that he is incompetent at John is incompetent b Does John knows that he is incompetent at John is incompetent c John doesn t know that he is incompetent at John is incompetent Entailments a John is French b Is John French c John isn t French gt John is European 75gt John is European 75gt John is European Why Study Presupposition I Presuppositions are ubiquitous John regrets that he is incompetent at John is incompetent John has stopped smoking at John used to smoke It is John who left at Someone left What John drank was vodka at John drank something She is clever at The person pointed at is female I John too was jailed 7 Someone other than John was jailed I John was jailed again at John was jailed before I Only John was jailed 7t Somebody was jailed Window to Paris Window to Paris Review by Marty Mapes In this Russian charmer the tenants of a Russian apartment building discover a magical wardrobe that opens onto the roof of a Paris apartment building The apartment dwellers desire to visit Paris and their reactions once there say more about the disparity in the standard of living than any political treatise If the lm s message is political it is not overtly so The characters are average people in search of a good time Their enthusiasm and sense of fun is contagious httpwwwmoviehabitcomreviewswinid00shtml Window to Paris Review by Marty Mapes In this Russian charmer the tenants of a Russian apartment building discover a magical wardrobe that opens onto the roof of a Paris apartment building The apartment dwellers desire to visit Paris and their reactions once there say more about the disparity in the standard of living than any political treatise If the lm s message is political it is not overtly so The characters are average people in search of a good time Their enthusiasm and sense of fun is contagious httpwwwmoviehabitcomreviewswinid00shtml Why Study Presupposition II Presuppositions and Dynamic Semantics Static View of Meaning Meaning Truth Conditions Dynamic View of Meaning after the 1980 s Meaning Context Change Potential potential to change beliefs Motivations for the dynamic View a Pronouns eg Every man who has a donkey beats it b Presuppositions Why Study Presupposition III The Semantics vs Pragmatics Divide I Semantics study of meaning as it is encoded in words John is an American student gt John is a student John is a former student 75gt John is a student I Pragmatics study of the additional information that can be obtained by reasoning on the speaker s motives Mr Smith is unfailingly polite and always on time gt Smith is a bad student Semantics vs Pragmatics 10 The rightmost person in the rst row is asleep rightmost the person in the rst row is asleep Credits httpwwwgiveusahomecouldfunfamoussnoopygif 0k 11 Semantics The rightmost person in the rst row is asleep True False 12 Semantics I Goal of Semantics Give systematic rules that derive the truth conditions of every sentence from a the meaning of its parts and b the way they are put together I Derived Notions a S is a tautology if S true in every conceivable situation b S is a contradiction if S is true in no conceivable situation 0 S entails S if every conceivable situation in which S is true is a situation in which S is true Test In every conceivable Situa t1390n in Which it is true that S it is true that S Note If S entails S S and not S is a contradiction 13 Pragmatics I Dear Colleague Mr John Smith has asked me to write a letter on his behalf Mr Smith is unfailingly polite is neatly dressed at all times and is always on time for his classes Sincerely yours Harry H Jones I Not an entailment The entailment test fails The following is no contradiction Mr Smith is unfailingly polite is neatly dressed at all times and is always on time for his classes But these are only his most super cial qualities Mr Smith is definitely a good student even an excellent one 14 Pragmatics Maybe I shouldn t hire Smith Smith is unfailingly polite and always on time 15 Pragmatics I An implicature i In a letter of recommendation the professor is normally supposed to mention the most positive features of the student ii Smith only mentioned that Smith is polite neatly dressed and always on time iii Therefore these are probably his most positive qualities and therefore he is probably a bad student 16 Pragmatics Utterance Can you pass the salt Implicature I want you to pass the salt A39s utterance Smith doesn39t seem to have a girlfriend these days B39s utterance He has been paying a lot of Visits to New York lately B39s implicature Smith has a girlfriend in New York 17 Entaihnents vs Implicatur es Difference 1 Entailments follow from what is linguistically encoded Implicatures do not Difference 2 Entailments satisfy the following test Implicatures generally don39t I To check whether p entails q check whether In every conceivable situation in which it is true that p it is true that q Difference 3 Implicatures can be cancelled Entailments cannot be 18 Scalar Implicatures a Rick is a philosopher or he is a poet B Schwarz b John will leave or Mary will leave 0 Paris is pleasant or London is pleasant l9 Hypothesis 1 Disj unction is unambiguously exclusive I i or i39 true iff exactly one of i i39 is true I Notational variant with 1 true 0 false i or i39 1 iff exactly one of i i39 is equal to 1 20 1 a is predicted to be a contradiction it should have the same status as b I a Rick is a philosopher or he is a poet In fact he is both b Rick is a philosopher or he is a poet but he is not both In fact he is both 2 Incorrect predictions I a Every Italian who is a philosopher or a poet is a socialist b Whenever I invite a philosopher or a poet to a party it ends up being a success 21 I Every Italian who is a philosopher or a poet is a socialist i1 is a philosopher but not a poet and he is a socialist i2 is a poet but not a philosopher and he is a socialist i3 is both a philosopher and a poet but he is not a socialist 22 Hypothesis 2 Disjunction is ambiguous l Crosslinguistic morphology 2 The ambiguity theory predicts that a could be understood as true in the situation we described earlier 3 Ellipsis Fox crediting T Stephenson I John read Chomsky or Montague Mary did too In fact she read both 23 General observation about ellipsis John went to the bank Mary did too banis ambiguous bankl slope near the side of a river bankz nancial institution Ok John went to the bankl Mary went to the bankl too Ok John went to the bankz Mary went to the bankz too John went to the bankl Mary went to the bankz too John went to the bankz Mary went to the bankl too Ok John went to the bankl Mary did ge te theb ank too Ok John went to the bankz Mary did ge te thebank too John went to the bank Mary did ge to thebank too John went to the bankz Mary did ge te theb anlg too 24 4 Yet another problem I a It is certain that John will read Chomsky or Montague b Every student read Chomsky or Montague 25 Hypothesis 3 Scalar Implicatures Hypothesisi or is inclusive disjunction ii an implicature is responsible for the not and inference S said F or G ltand orgt form a scale F and Gentails F or G If S believed that F and G it would have been more cooperative to say F and G Primary Implicature NOT S believes F and G If John is well informed and either believes or disbelieves F and G we also get Secondary Implicature S believes NOTF and G 26 I Alternatives I AltS S39 S39 is a sentence obtained from S by replacing simultaneously any number of occurrences of or by and and any number of occurrences of and by or I a 81 Rick is a philosopher or a poet AltSl Rick is a philosopher or a poet Rick is a philosopher and a poet b 82 Rick is a philosopher and a poet AltSz AltSl Rick is a philosopher or a poet Rick is a philosopher and a poet c S3 I doubt that Rick is a philosopher and a poet AltS3I doubt that Rick is a philosopher and a poet I doubt that Rick is a philosopher or a poet 27 II Ordering and Cooperation I Ordering Let S be a sentence and let S39 be a member of AltS S39 is better than S if a S39 entails S and S does not entail S39 terminologyz we say that S asymetrically entails S b The speaker believes that S39 I Cooperation A sentence S is not uttered cooperatively if for some S39 in AltS S39 is better than S 28 Scalar Implicatures I a Rick is a philosopher or a poet b AltaRick is a philosopher or a poet Rick is a philosopher and a poet cand gtgtor a is not uttered cooperatively if the speaker believes that Rick is a philosopher and a poet Primary Implicature If the speaker is cooperative it39s not the case that the speaker believes that Rick is both a philosopher and a poet Secondary Implicature If the speaker has an opinion on this matter it must be that he believes that Rick is not both a philosopher and poet 29 Scalar Implicatures a Rick is a philosopher and a poet b AltaRick is a philosopher and a poet Rick is a philosopher or a poet 0 No member of Alta asymmetrically entails a so nothing additional is inferred 3O 39 Scale Rever sal39 I a I doubt that Rick is a philosopher and a poet b AltaI doubt that Rick is a philosopher or a poet I doubt Rick is a philosopher and a poet c I doubt that or gtgt I doubt that and a is not uttered cooperatively if the speaker doubts that Rick is a philosopher or a poet hence if the speaker is cooperative the speaker does not doubt that Rick is a philosopher or a poet ie he believes that Rick is a philosopher or a poet a philosopher and poet 31 39 Scale Rever sal39 a Every Italian who is a philosopher or a poet is a socialist gt no additional inference because the version with and would be less informative b Every Italian who is a philosopher and a poet is a socialist gt it s not the case that every Italian who is a philosopher or a poet is a socialist ie sorne Italian who is a philosopher or a poet but not both is not a socialist 32 39 Scale Rever sal39 I a Whenever John is next to Mary or Ann he behaves like an idiot gt no additional inference b Whenever John is next to Mary and Ann he behaves like an idiot gt It39s not the case that Whenever John is next to Mary or Ann he behaves like an idiot 33 Some Most Every I a Some of my friends are clever gt Not all of my friends are clever gt A minority of my friends are clever b Some of my friends are clever In fact all of them are I a Most of my friends are clever gt Not all of my friends are clever b Most of my friends are clever In fact all of them are I a Whenever most of the students come to class there is a pleasant atmosphere b Every student who read most of the articles on the reading list will get an A 34 Extensions I ltand orgt ltall most somegt ltcertain probablelikely possiblegt lt 6 5 4 3 21gt ltb0iling hot warmgt ltad0re love likegt ltexcellent good 0kaygt 35 Why are Scales Necessary I a John read some book b John read exactly one book 0 b is more informative than a therefore the speaker was not in a position to assert b d Therefore it is likely that John didn t read exactly one book This is the opposite of the result we want 36 Other Implicatures John is in Paris or he is in Rome gt it is not the case that a the speaker believes that John is in Paris b the speaker believes that John is not in Paris 0 the speaker believes that John is in Rome 1 the speaker believes that John is not in Rome If John is in Paris he is there for business gt the speaker takes it to be possible but not certain that John is in Paris 37 Experiment Scalar Implicatures Crain amp coworker s U Maryland 38 Test sentence Every spaceguy took a strawberry or an onion ring 39 Creditsz Crain amp coworkers U Maryland 40 I d like to take this banana but it won t go through the door Creditsz Crain amp coworkers U Maryland 41 I am just gonna take this Creditsz Crain amp coworkers U Maryland 42 Creditsz Crain amp coworkers U Maryland 43 Everr spaee guy took a strawberry or an anion ring 15 Englishspeaking Children age 34 to 62 mean 572 accepted the target sentence 3 tunes out of 60 tnals 50 Control Home I 8 adults alwavs re1 ected the target sentences Creditsz Crain amp coworkers U Maryland 44 Children and Scalar Implicatures I Children appear not to compute Scalar Implicatures in some environments Where adults do I Paradox children appear to be 39more logical39 than adults 45 Exiery space guy took a strawberry and an onion ring Every Space guy took a strawberry or an onion ring 46 Every spaceguy took a strawberry g an onion ring Every spacegny took a strawberry and an onion ring 15 children age from 32 to 60 mean 4 rewarded only the puppet who had used the conjunction and in 56 cases out of 60 trials 933913 47 Scalar Implicatur es Take Time Noveck and Posada 2003 Patently True P atently False Underinfbrmative Some people have brothers Some animals have stripes Some heuses have bricks SUIT CO L ChCS have windows 30 me ea is have parents Seine lmngames have airplanes Some turtles have shells Some gira es have neeks Some sentences have wmds 48 n f spauise m Paf n39tl i smemem 13ateintly False siaremam39 dinLughnmnnm l l m Thus whn Itspond logi cale m the Und lrinfmm rive 511111an En Tham whn 1E5de prag matically m The indium fmma ve awaitmam n 2 1 3 fatal 645 633 356 TM Unde mm 655 1203 1014 49 Three Properties of Scalar Implicatures Unlike entailments they can be cancelled They disappear in certain environments and appear in others They are acquired relatively late by children They take time to compute 50 Pr esuppo sitions 51 Pr esuppositions vs Entailments Difference 1 dubious If an entailment of S is false S is false not weird John is French No He is South African John knows that he is going to be red No He doesn t know it ltgt No He is going to keep his job 52 Pr esuppositions vs Entailments Difference 2 very clear Presuppositions project differently from entailments a Is John French 75gt John is European b John is not French 75gt John is European c None of these 10 students is French 75gt Each of these 10 students is European 75gt Some of these 10 students is European a Does John know that he is incompetent gt John is incompetent b John does not know that he is incompetent gt John is incompetent c None of these 10 students knows that he is incompetent gt Each of these 10 students is incompetent 53 Pr esuppositions vs Entailments I a Does John take care of his computer gt John has a computer b John doesn t take care of his computer gt John has a computer c None of these 10 students takes care of his computer gt Each of these 10 students has a computer I 21 Did John stop smoking gt John used to smoke b John didn t stop smoking gt John used to smoke c None of these 10 students stopped smoking gt Each of these 10 students used to smoke 54 Pr esuppositions vs Implicatur es I An analysis of presuppositions as implicatur es Hypothesis If pp is a clause described as having presupposition p and assertion p i pp has as its meaning the conjunction of p and p ii but ltp pp gt forms a scale I Examples a ltIt is raining John knows that it is raininggt b ltJohn smoked John has stopped smokinggt c ltJohn has a girlfriend John loves his girlfriendgt 55 Predictions I 3 pp entails p ltOkgt a John knows that he is incompetent gt John is incompetent b I ll inVite John and Mary gt I ll invite John or Mary not pp implicates p ltOkgt because not p is more informative than not pp a John doesn t know that he is incompetent implicates John is incompetent b I won t inVite both John and Mary gt I ll inVite John or Mary 56 Predictions 11 I No student PP implicates Some student not P because No student P is more informative than No student PP hence the inference that not No student P ie Some student not P I lt3 students PP implicates 23 students not P because Less than 3 students P is more informative than Less than 3 students PP hence the inference that not Less than 3 students P ie At least 3 students not P These are the crucial cases to test 57 Pr esuppositions vs Entailments An Experiment French Chemla 2007 in EE native speakEra df39 French CdntEext Agataquot um Hiram SESS39iiG 1 539 aquot 37 Ema m s indal widua y armEit IE tudu nm Elf theair cid g airmiud y 5 5m dEnt f d Jahnij HEESIE teachers mam i f 39 quot f y discus5 Erbium themquot S d t T eae teaShem an quotvery 39HJEM ima fdaquot39med d dut theii studentgy h 39 l s faila1 Ndn l gil l inferential hawk 2 Eii l pIEE were 13139dvidEsd Jth E I39vf I rgr39 did 5L JDhIi did A and Bi SMQQEEES Mimi Suggnggtg ia Jim1111 did A th did A heftre ill PM litres Nd res mite Eliminated pdt ti l prdbIEIILE due tn ddmai restric tidns 133 EIl liEitIF referring 1213 a particular 5E1 f individu als egg 1 finFame df these Iii S HdETHES I Epl C l Na 5mderii 58 Experimental Conditions I Triggers Presuppositions attitude verbs know be unaware change of state start stop de nite descriptions his computer Implicatures ltall somegt ltand orgt ltexcellent goodgt I Environments Inferences universallike and implicaturelike Operators John I doubt that John More than 3 of these 10 students Each of the 10 students None of these 10 students Exactly 3 of these 10 students 59 Examples Less than 3 of these 10 students know that their father is about to receive a congratulation letter gt The father of each of these students is about to receive a congratulation letter gt The father of at least 3 students is about to receive a congratulation letter None of these 10 students read the handout and did an exercise gt Each of these 10 students did at least one or the other gt At least 1 of these 10 students did at least one or the other 60 Main Results Chemla 2007 I Presuppositions display a different a behavior from scalar implicatures under 110 Nonuniversal inferences for implicatures Universal implicatures for presuppositions I Not all quantifiers behave on a par at least 3 more than 3 eXactly3 display an intermediate behavior universal inferences half the time I Not computing a presupposition takes time 61 40 60 80 100 20 N0 and Universal Infer ences Pres El EachgtEach I NogtAtLeastOne l NogtEach Impl 62 40 60 80 100 20 Less than three and Univer sal Infer ences Pres III EachgtEach l LessThan3gtAtLeastE l LessThan3gtEach Impl 63 Universal Inferences for Various Quanti ers Each No Less than 3 Mare than 3 Exactly 3 100 EU I 4D 2D Reaction Times Universal Inferences Response times ms 6000 000 8000 9000 10000 5000 I YES responses I NC responses PRESUPPOSITIONS I MPLICATURES 65 Questions I Triggering Problem Why do some elementary clauses have presuppositions a John knows that it is raining 7 It is raining b John rightly believes that it is raining 7t none or possibly John believes that it is raining 66 Questions I Projection Problem How do the presuppositions of elementary clauses get transmitted to complex clauses a If John is realistic he knows that he is incompetent at John is incompetent b If John is an idiot he knows that he incompetent 7t none or possibly if John is an idiot he is incompetent 67 Questions I Architectural Question Where do presuppositions belong in the architecture or language Are they a semantic or a pragmatic phenomenon 68 The Projection Problem 69 I a Is it true that John knows that he is incompetent John is incompetent I doubt that John knows that he is incompetent John is incompetent None of these 10 students knows that he is incompetent Each of these 10 students is incompetent a Cl a C a CT aaoacrm Conjunction John knows that he is incompetent John is incompetent and knows that he is Is it true that John is incompetent and knows that he is HODC I doubt that John is incompetent and knows that he is none None of these 10 students is incompetent and knows it none 70 Conjunction I a John is depressed and his boss knows that he is incompetent b Is it true that John is depressed and that his boss knows that he is incompetent at John is incompetent c I doubt that John is depressed and that his boss knows that he is incompetent I a John is an idiot and his boss knows that he is incompetent b Is it true that John is an idiot and that his boss knows that he incompetent 7t if John is an idiot he is incompetent C c I doubt that John is an idiot and that his boss knows that he is incompetent 71 Conjunction p and gq presupposes p 2 q to be re ned John is incompetent and he knows it that he is nznone John is an idiot and he knows that he is incompetent 7t if John is an idiot he is incompetent John is depressed and his boss knows that he is incompetent Predicted 71 If John is depressed he is incompetent Actual at John is incompetent Maybe because the most plausible way to make the conditional true is to assume that its consequent is 72 Conditionals I a If John is incompetent he knows that he is b Is it true that if John is incompetent he knows that he is c I doubt that if John is incompetent he knows that he is I a If John is realistic he knows that he is incompetent b Is it true that if John is realistic he knows that he is incompetent c I doubt that if John is realistic he knows that he is incompetent I a If John is over 65 he knows he can t apply b Is it true that if John is over 65 he knows he can t apply c I doubt that if John is over 65 he knows he can t apply 73 Conditionals I a If John knows that he is overquali ed he won t apply b Is it true that if John knows that he is overquali ed he won t apply 0 I doubt that if John knows that he is overquali ed he won t apply I a If John knows that he is overquali ed he is depressed b Is it true that if John knows that he is overquali ed he is depressed c I doubt that if John knows that he is overquali ed he is depressed I a ifp gq presupposes p gt q b ifpp q presupposes p 74 Disj unctions a If John is incompetent he knows that he is b Either John is not incompetent or he knows that he is a If John is realistic he knows that he is incompetent bEither John is not realisticor he knows he is incompetent a If John is over 65 he knows he can t apply b Either John isn t over 65 or he knows he can t apply a If John knows that he is overquali ed he won t apply b Either John doesn t know that he is over quali ed or he won t apply a p or gq presupposes not p 2 q b pp or q presupposes p 75 Stalnaker s Pragmatic Analysis 76 A Pragmatic Analysis I I p and gq9 presupposes p Z q when a speaker says something of the form A and B he may take it for granted that A or at least that his audience recognizes that he accepts that A after he has said it The proposition that Awill be added to the background of common assumptions before the speaker asserts that B Now suppose that B expresses a proposition that would for some reason be inappropriate to assert except in a context where A or something entailed by A is presupposed Even if A is not presupposed initially one may still assert A and B since by the time one gets to saying that B the context has shifted and it is by then pr esupposed that A Stalnaker Pragmatic Presuppositions 1974 77 Assumptions I Assumption 1 Sentences may be true false or I Assumption 2 A sentence S is a presupposition failure if it has the value with respect to atleast one of the states of affairs compatible with what the speech act participants take for granted De nition 1 Common Ground what the speech act participants take for granted De nition 2 Context Set set of worlds compatible with what the speech act participants take for granted I Assumption 3 The Context Set is updated incrementally in discourse and in conjunctions 78 I A possible world W a complete specification of what is Possible Worlds going on It determines for every sentence S whether S Wtrue S Wfalse01 S w I Different clauses give rise to different functions eg The President of France is Chirac WI gt false w2 gt true W3 gt W4 gt The US President is Bush WI gt true w2 gt false W3 gt true W4 gt Two plus two is four WI gt true w2 gt true W3 gt true W4 gt true 79 Further Conditions I NonContradiction A sentence S uttered in a Context Set C is deviant if S is true in no world of C I NonTriviality A sentence S uttered in a Context Set C is deviant if S is true in every world of C 80 Stalnaker 3 Analysis I John knows thatlle 1395 Incompetent is true in W if John is incompetent and believes that he is false in W if John is incompetent and doesn t believe he is in W if John is not incompetent I Suppose that the speech act participants do not know Whether John is or isn t incompetent Suppose further that the Context Set C is C W1 W2 W3 W4 W1 John is incompetent and believes that he is W22 John is incompetent and believes he isn t W3 John is not incompetent but believes he is W4 John is not incompetent and believes he isn t 81 Stalnaker 3 Analysis I T John knows thatlle is Incompetent uttered in C is a presupposition failure because this sentence is in W3 and W4 which both belong to C I Suppose that the speech act participants do not know Whether John is or isn t incompetent Suppose further that the Context Set C is C W1 W2 W3 W4 W1 John is incompetent and believes that he is W22 John is incompetent and believes he isn t W3 John is not incompetent but believes he is W4 John is not incompetent and believes he isn t 82 Stalnaker 3 Analysis I S John is Incompetentis true in w if John is incompetent in W false in win all other cases ie the sentence does not have a presupposition I a Acceptability Clearly John is Incompetent uttered in C is not a presupposition failure b Update Initially the Context Set was C w1 w2 w3 w4 After S is uttered the new Context Set is C w1 w2 ie only the worlds compatible with S are retained 83 Stalnaker 3 Analysis John is incompetent He knows it S T Step 1 The initial Context Set is C w1 w2 w3 w4 After the first sentence is uttered the new Context Set is C w1 w2 Step 2 The second sentence is evaluated with respect to C By construction in each world in C T has a value different from So T is not a presupposition failure in C Step 3 C is updated to C W1 84 Stalnaker s Analysis I Conjunction a Treat S and Tin the same way as the discourse S T the assertion of a conjunction is a succession of two assertions b Beautiful analysis of presupposition projection every world in C that satis es S must satisfy T In other words C S gt T l Limitations a How does the analysis extend to other operators b How does the analysis extend to embedded conjunctions e g None Ofmy students is rich andproua 01 85 Heim s Semantic Analysis following in part Karttunen 1974 86 Heim s Analysis I a Keep from Stalnaker s analysis the idea of an update the analysis of presupposition projection in conjunctions b Drop the pragmatic derivation of Stalnaker s analysis c Result Old conception meanings as truth conditions New conception meanings as Context Change Potentials ie as functions from Context Sets to Context Sets I Notations CF the result of updating the Context Set C with F 87 Heim s Analysis I Elementary Clauses 21 CJohn is incompetent iff C we C John is incompetent in w otherwise b CJohn knows that he is incompetent iff C or for some weC John is not incompetent in w we C John believes he is incompetent in w otherwise I Conjunction CF and G CFG 88 Heim s Analysis I Negation Cnot F iffCF C CF otherwise 3 F I a not F John doesn t know that he is incompetent b Cnot F iffCF iff for some weC John is not incompetent in W C CF otherwise ie C we C John believes he is incompetent in w 89 Heim s Analysis I Negation In nzzf39 This means that Cnot F iffCF notpp C CF OtherWiSG 3presupposes that p D F I a not F John doesn t know that he is incompetent b Cnot F iffCF iff for some W6C John is not incompetent in W C CF otherwise ie C we C John believes he is incompetent in W 9O Heim s Analysis I Conditionals analyzed as material implications CifF G iffCF or CFnot G C CFnot G otherwise Worlds that refute 39 if F G F G 91 Heim s Analysis I Conditionals analyzed as material implications CifF G iffCF or CFnot G C CFnot G otherwi sequotquotquot quot This means that if pp q presupposes that p and that if p gq presupposes if p q t I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I l I I Worlds that refute 39 if F G F G 92 Heim s Analysis I if F G If John is incompetent he knows it I CifF G iffCF or CFnot G But CF 7E and furthermore CF WE C John is incompetent in W CFnot G iff CFG which is not the case by construction Furthermore CFnot G WE C John is incompetent in W not G WE C John is incompetent in W WE C John is incompetent in W and John believes he is incompetent in W WE C John is incompetent but doesn t believe it in W Cif F G C WE C John is incompetent but doesn t believe it in W 93 Summary Meaning of an elementary clause a CCP Conjunction CF and G CFG Negation Cnot F iffCF C CF otherwise Conditionals CifF G iffCF or CFnot G C CFnot G otherwise Disjunction CF or G iffCF or Cnot FG CF U Cnot FG otherwise 94 Disj unctions a If John is incompetent he knows that he is b Either John is not incompetent or he knows that he is a If John is realistic he knows that he is incompetent bEither John is not realisticor he knows he is incompetent a If John is over 65 he knows he can t apply b Either John isn t over 65 or he knows he can t apply a If John knows that he is overquali ed he won t apply b Either John doesn t know that he is over quali ed or he won t apply a p or gq presupposes not p 2 q b pp or q presupposes p 95 Heim s Analysis I Disj unction CF or G iffCF or Cnot FG CF U Cnot FG I F G I a John is not incompetent or he knows that he is b Cnot I or K iffCnot I or Cnot not IK ie iff CI or CIK which is never the case Thus Cnot I or K Cnot I U CIK 96 Heim s Analysis I Disjunction CF 01 G iffCF or Cnot FG CF U Cnot FG 3 y This means that quot pp or q pre supposes that p and that p or I gq presupposes F G kif not 13 q j 9 O I I I I a John is not incompetent or he knows that he is b Cnot I or K iffCnot I or Cnot not IK ie iff CI or CIK which is never the case Thus Cnot I or K Cnot I U CIK 97 Heim s Analysis I De nition of Truth If WE C a F is in W relative to C iffCF b H75 F is true in W relative to C iffvveCF I John is incompetent He knows it S T C 2 W1 W2 W3 W4gt CS C W13 W2 CST C W1 I a Relative to W1 C the discourse is true since W16 CST b Relative to W2 C the discourse is false since 98 W2 CST 99 Heim s Explanatory Problem I Problem is the account explanatory Soames 1989 CF and G CFG CF andquotlt G CGF When F and G are not presuppositional CF and GCF and GWE C F is true in W and G is true in W 100 Heim s Explanatory Problem I There are many ways to de ne the CCP of or CF or1 G CF U CG unless one ofthose is CF or2 G CF U Cnot FG unless one of those is CF or3 G Cnot G F U CG unless one of those is lOl Gazdar s Account 102 An Explanatory Account Step 1 Compute the various implicatures of a sentence Step 2 Keep only those presuppositions that are consistent with all implicatures John is incompetent and he knows that he is a Implicature If John is Incompetentis uttered it cannot be trivial that John is incompetent ie C 7 John is incompetent b Potential Pr esupposition the second conjunct triggers the potential presupposition that John is Incompetent c Filtering The presupposition is filter ed out because it is inconsistent with the implicatur e 103 An Explanatory Account I John is depressed and he knows that he is incompetent I a Implicatur e If John is depressed is uttered it cannot be trivial that John is depressed ie C 7 John is incompetent b Potential Pr esupposition the second conjunct triggers the potential presupposition that John is Incompetent c Filtering The presupposition is not ltered out because it is consistent with the implicature Note Gazdar thus predicts that the entire sentence presupposes that John is depressed Stalnaker and Heim predict I39fJOlm is depressed he is Incompetent Most examples go in Gazdar s direction 104 An Explanatory Account I If John is incompetent he knows it I a Implicature The speaker cannot utter 1H7 G felicitously if he knows that F is true If we represent as S the set of worlds compatible with what the speaker believes S 7 John is incompetent from which it follows that C 7 John is incompetent b Potential Pr esupposition the main clause triggers the potential presupposition that John is Incompetent c Filtering The presupposition is lter ed out because it is inconsistent with the implicatur e 105 Reminder Other Implicatures John is in Paris or he is in Rome gt it is not the case that a the speaker believes that John is in Paris b the speaker believes that John is not in Paris 0 the speaker believes that John is in Rome 1 the speaker believes that John is not in Rome If John is in Paris he is there for business gt the speaker takes it to be possible but not certain that John is in Paris 106 An Explanatory Account I Either John is not incompetent or he knows that he is a Implicature The speaker cannot utter F or G felicitously if he believes that F is false S 7 John is incompetent from which it follows that C 7 John is incompetent b Potential Pr esupposition the second clause triggers the potential presupposition that John is incompetent c Filtering The presupposition is lter ed out because it is inconsistent with the implicatur e Note Gazdar thus predicts that the entire sentence presupposes that John is depressed Stalnaker and Heim predict ifJOnn is depressed lie is incompetent 107 An Explanatory Account I If John is depressed he knows that he is incompetent I a Implicature S 7E John is depressed from which it follows that C 7 John is depressed b Potential Pr esupposition the main clause triggers the potential presupposition that John is Incompetent c Filtering The presupposition is not ltered out because it is consistent with the implicature Note Gazdar thus predicts that the entire sentence presupposes that John is depressed Stalnaker and Heim predict I39fJOlm is depressed he is Incompetent Most examples go in Gazdar s direction but not all do 108 Problem for Gazdar s Account I If John is French he must know that he can travel within the European Union without a passport a Gazdar s prediction at John can travel within the European Union without a passport b Actual presupposition probably none I a Implicature S 7 John is French from which it follows that C 7 John is French b Potential Pr esupposition the main clause triggers the potential presupposition that John can travel Within the European Union Without a passport 0 Filtering The presupposition is not ltered out because it is consistent with the implicature I If John has twins then Mary will not like his children 109 Back to Heim s Account Quanti cation not discussed in lectures much more technical content 110 Replacing Worlds with Contexts Examplel An amnesiac gets lost An amnesiac Rudolf Lingens is lost in the Stanford library He reads a number of things in the library including a biography of himself and a detailed account of the library in which he is lost He still won t know who he is and where he is no matter how much knowledge he piles up until that moment when he is ready to say 172139s place is aisle ve oor siX of Main Library Stanford I am Rudolf Lingens Perry 1977 It seems that the Stanford library has plenty of books but no helpful little maps with a dot marked location of this map Book learning will help Lingens locate himself in logical space But none of this by itself can guarantee that he knows where in the world he is He needs to locate himself not only in logical space but also in ordinary space Lewis 1979 p 138 111 Example 2 39My pants are on re39 I If I see re ected in a Window the image of a man Whose pants appear to be on re my behavior is sensitive to Whether I think His pants are on re or My pants are on re though the object of thought may be the same39 Kaplan 114 115 Refer ential Uncertainty I Situation Lingens who is lost in the Stanford library knows everything there is to know about the world I wear a coat My coat is black I Lingens a wellread amnesiac knows everything there is to know about the world but he does not know whether he is Alfred who is having a conversation with Bereniee or Charles who is having a conversation with Denise Berenice used to smoke but Denise never did Compare Did you stop smoking You are Berenice Did you stop smoking 116 Refer ential Uncertainty I Situation Lingens who is lost in the Stanford library knows everything there is to know about the world I wear a coat My coat is black I Lingens a wellread amnesiac knows everything there is to know about the world but he does not know whether he is Alfred who is pointing towards Bereniee or Charles who is pointing towards Denise Berenice used to smoke but Denise never did Compare Did she stop smoking She is Berenice Did she stop smoking 117 Static Account with worlds I it is raining W false PS is in St Petersburg W true the British President is happy W I Rule Pro VP W true if and only if Pro W 6 VP W Where VP W is the set of things of which VP is true in W Pro VP W false if and only if Pro W 6 VP 39 W Where VP H W is the set of things of which VP is false in W Pro VP W in all other cases 118 Static Account with contexts A context ltWorld speaker addressee denotation of prol gt 1 smoke ltW PS YOU John Mary gt true if and only if PS smokes in W false if and only PS does not smoke in W Shez smokes ltW PS YOU John Mary gt true if and only if Mary smokes in W false if and only Mary does not smoke in W Shez stopped smoking ltW PS YOU JOhn Mary 39gt true if and only if Mary used to smoke but doesn t nOW in W false if and only Mary used to smoke and still does in W if and only if Mary didn t use to smoke 119 Dynamic Account with contexts The rules are exactly the same as before replacing worlds with contexts Elementary Clauses now C is a set of contexts We write as cW the world of c as Cl the denotation of pI39Ol a CJohn is incompetent iff C CE C John is incompetent in cw otherwise b CJohn knows that he is incompetent iff C or for some ceC John is not incompetent in cW CE C John believes he is incompetent in cw otherwise 0 Cshe2 stopped smoking iff C or for some ceC 02 did smoke in cW CE C 02 doesn t smoke in cw otherwise 120 Quanti cation in a Static Setting lt PS YOU J h M I 110 X13 X1 StUant X1 smokes W 9 a 0 r1 ary gt 75 iff for every d which is a student in w Xl smokes ltw PS YOU 1 Mary gt i If 75 true iff for no d which is a student in w lt PS YOU 1 M gt X1 smokes W ary true false iff for some d which is a student in w ltw PS YOU 1 Mary gt X1 smokes true 121 122 Quanti cation in a Static Setting I no X1 X1 student X1 stopped smokingltwj PS YOU Johngt 7f iff for every d which is a student in w X1 stopped Smoking ltW PS YOU d Mary gt If 75 true iff for no d which is a student in w ltw PS YOU 1 Mary gt X1 stopped smoking true false iff for some d which is a student in w X1 stopped smoking ltW PS YOU d Mary quot39gt true 123 Quanti cation in a Dynamic Setting I Notations ci gt d that context which is exactly like 0 except that proi denotes d Ci gt d ci gt 1 06C I CIIO Xi Xi NP XiVP lff C or ci gtd 06C and dis an objectxi NP or ci gtd 06C and ci gtd e Ci gtdxi NP xi VP H75 Cno xi xi NP xi VP 0 06C and for no object 1 ci gtd e Ci gtdxi NP and ci gtd e Ci gtdxi NPxi VP I Cevery Xi Xi NP xiVP same thing as for no replacing no with every 124 Quanti cation in a Dynamic Setting I no X1 x1 student x1 smokes Let us assume that C 75 Then Cno X1 X1 student X1 smokes 7E because C c1 gtd 06C and dis an object X1 student 75 and c1 gtd 06C and c1 gtd e C1 gtdX1 student X1 smokes 7E Furthermore Cno X1 X1 student X1 smokes 0 06C and for no object d c1 gtd e C1 gtdX1 student and c1 gtd e C1 gtd X1 student X1 smokes 0 06C and for no object d dis a student in cW and d smokes in cw 125 Quanti cation in a Dynamic Setting I no X1 x1 student x1 smokes Let us assume that C 01 Cg c3 c4 and for each i ci ltWi PS YOU Johngt with W1 All students used to smoke All students still smoke W22 All students used to smoke One doesn t any more W3 One student didn t use to smoke No student smokes W4 One student didn t use to smoke One student smokes Cno X1 X1 student X1 smokes ciz i e 1 2 3 4 and for no object d ltWi PS YOU dgt e ltWi PS YOU dgt i e 1 2 3 4X1 student and ltWi PS YOU dgt E ltWi PS YOU dgt i e 1 2 3 4X1 student X1 smokes ciz i e 1 2 3 4 and for no object d dis a student in Wi and d smokes in Wi 03 126 Quanti cation in a Dynamic Setting I no X1 x1 student x1 stopped smoking Let us assume that C 75 Then Cno X1 X1 student X1 smokes iff c1 gtd 06C and dis an ObjCCtX1 student or c1 gtd 06C and c1 gtd e C1 gtdX1 student X1 stopped smoking iff c1 gtd 06C and c1 gtd e C1 gtdX1 student X1 stopped smoking iff for each ceC if d is a student in cw d used to smoke in cw If 75 0 06C and for no object d dis a student in cW and d stopped smoking in cw 127 Quanti cation in a Dynamic Setting I no xi xi student xi stopped smoking Let us assume that C 01 02 c3 04 with cl 02 c3 04 de ned as before Cno X1 X1 student X1 stopped smoking because c1 gtd 06C and c1 gtd e C1 gtdX1 student X1 stopped smoking ltWi PS YOU dgt i e 1 2 3 4 and dis a student in Wi X1 stopped smoking because in W3 and W4 there are students who didn t use to smoke 128 Quanti cation in a Dynamic Setting I no xi xi student xi stopped smoking Let us now assume that C 01 02 with 01 and 02 de ned as before It can be shown C no X1 X1 student X1 smokes 7E Furthermore C no X1 X1 student X1 stopped smoking 0 CEC and for no object d dis a student in cW and d stopped smoking in CW 2 Cl 129 Back to Heim s Account Accommodation 130 Global Accommodation I My sister is pregnant I 39 it39s not as easy as you might think to say something that will be unacceptable for lack of required presuppositions Say something that requires a missing presupposition and straightway that presupposition springs into existence making what you said acceptable after all I said that presupposition evolves in a more or less rulegoverned way during a conversation Now we can formulate one important governing rule call it the Rule of accommodation for presupposition If at time t something is said that requires presupposition P to be acceptable and if P is not presupposed just before t then ceteris paribus and within certain limits presupposition P comes into existence at tquot 131 Local Accommodation I a The king of France is not Wise because there is no king of France b None of my students takes good care of his car because none of my students has a car 0 John doesn39t know that he is incompetent because he just isn39t incompetent I a It39s not the case there is a king of France and he is Wise because b None of my students has a car and takes good care of it because 0 It39s not the case that John is incompetent and knows it because 132 Global vs Local Accommodation I Cnot F iffCF C CF otherwise I Global Accommodation C39 06 C France is a monarchy at the time and in the world of c We then compute C39the king of France is not powerful I Local Accommodation Instead of computing C CF which wouldn39t even be de ned since CF we compute C C39F where C39ceC France is a monarchy at the time and in the world of c as in A 133 Directions I Allow for local accommodation whenever global accommodation would contradict a the literal meaning of a sentence b or an implicature of a sentence or possibly certain types of implicatures eg primary implicatures I In effect this allows us to capture the good properties of Gazdar s system within Heim s dynamic semantics 134 Summary I Presuppositions cannot be analyzed as implicatures I The dilemma of dynamic semantics a Stalnaker s approach is explanatory but not general Update the context set in time as you process a sentence b Heim s approach is general but not explanatory The meaning of words is dynamic from the start ie their lexical entries specify how they change the context set I Gazdar s account was explanatory and general but incorrect i Compute the implicatures of a sentence ii Project those potential presuppositions that don t contradict the entire sentence or one of its implicatures 135 The Proviso Problem I a If the problem was easy it is not John who solved it b John knows that if the problem was easy someone solved it Geurts 1999 I Predicted presupposition of a and b If the problem was easy someone solved it Actual presupposition of a Someone solved the problem Actual presupposition of b If the problem was easy someone solved it 136 The Proviso Problem John is an idiot and he knows that he is incompetent 7t if John is an idiot he is incompetent John is depressed and he knows that he is incompetent Predicted at If John is depressed he is incompetent Actual at John is incompetent Maybe this is because the most plausible way to make the conditional true is to assume that its consequent is but this kind of reasoning fails to address the minimal difference between If the problem was easy it is not John who solved it John knows that if the problem was easy someone solved it Geurts 1999 137 The Proviso Problem I Direction 1 van der Sandt 1992 Geurts 1999 This problem refutes the standard dynamic approaches as well as all approaches that make similar predictions A different analysis must be proposed in which presuppositions are treated in a more syntactic fashion Discourse Representation Theory This is a major contender among current theories I Direction 2 still promissory With enough pragmatic reasoning we can stick to Heim s predictions which in any event seem to be correct in other cases eg If John is over 65 he must know that he is too old to apply 138 The Transparency Theory 139 Plan I Lessons of dynamic semantics to be disputed a Sentences can be true false or b Trivalent logic alone doesn t suffice the logic must be dynamic I Claims a The dynamic approach is caught in a dilemma Stalnaker s approach is explanatory but not general Heim s approach is general but not explanatory b Gazdar s approach is explanatory and general but false 0 An analysis which is explanatory general and roughly empirical adequate can be obtained by deriving presuppositions from implicatures of manner 140 Context Update I Pragmatics I Stalnaker39s Analysis a pragmatic solution a John is incompetent and he knows that he is Step 1 Update the Context Set C with J is incompetent CJohn is incompetentwe C J is incompetent in wC39 Step 2 Update the intermediate Context Set C with he knows that lie is incompetent C39he knows itwe C J is incompetent in w and J believes in w that J is incompetent b John knows that he is incompetent and he is I Ideas i Assertp and q Assertp Assertq ii Pragmatic analysis 141 Context Update I Pragmatics I Problems with Stalnaker39s Analysis a It is not clear that the notion of 39intermediate Context Set39 makes sense e g None Ofmy students is both rich andproua 01 b It is unclear how the analysis can extend say to disjunction or quanti ers eg a disjunction cannot be equated with a succession of two assertions c Why should one update the Context Set anyway 142 Context Update I Pragmatics when a speaker says something of the form A and B he may take it for granted that A or at least that his audience recognizes that he accepts that A after he has said it The proposition that Awill be added to the background of common assumptions before the speaker asserts that B Now suppose that B expresses a proposition that would for some reason be inappropriate to assert except in a context Where A or something entailed by A is presupposed Even if A is not presupposed initially one may still assert A and B since by the time one gets to saying that B the context has shifted and it is by then pr esupposed that A Stalnaker Pragmatic Presuppositions 1974 143 Context Update II Semantics I Heim39s Analysis semantic reinterpretation of Stalnaker s rules a Rule CF and G C FG unless CF b Results same as before except that they can be extended to other connectives to quanti ers I Heim39s Explanatory Problem CF and G CFG CF andquotlt G CGF 144 Dynamic Semantics vs Transparency I Dynamic Semantics lhar dcore semanticsl a Trivalent dynamic b Insuf ciently predictive CF and G CFG CF andquotlt G CGF I Transparency Theory gt lmanner implicaturesl a Bivalent nondynamic b Manner Principle 1 Violable Be Articulate requires that you should say p and pp rather than pp c Manner Principle 2 Eviolable LeastE 39ortrules out a conjunction that starts with p and if p and is certain to do no work no matter how the sentence ends 145 The Intuition I Example 1 John knows that it s raining Be Articulate If possible say It s raining and John knows it rather than John knows that it s raining Least Effort Do not say It s raining and if for all p C it s raining and p ltgt p which happens if Presupposition C it s raining 146 The Intuition I Example 2 If it s raining John knows it Be Articulate If possible say If it s raining it s raining and John knows it rather than If it s raining John knows it Least Effort Do not say If it s raining it s raining and if for all p C if it s raining it s raining and p cgt if it s raining p which is always true Pr esupposition None 147 Two Principles of Manner I Be Articulate prossible say p and pp rather than pp eg Say John is incompetent and he knows it that he is I Least Effort Do not say 0 andg a rather than g a Version 1 Incremental if for each 0 of the same type as d and for each acceptable sentence completion 1 C adandc b ltgtac b eg If John is incompetent he is incompetent and sad Two Principles of Manner I Be Articulate prossible say p and pp rather than pp eg Say John is incompetent and he knows it that he is I Least Effort Do not say 0 andg a rather than g a Version 2 Global if for each c of the same type as d and fer eaeh C adandc b ltgtac b eg John is incompetent and sad if he is incompetent Incremental Transparency I Least Effort Let g be of type t or lte tgt If for each c of the same type as d and for each acceptable sentence completion b C adandc b ltgtac b d and should not have been uttered in the rst place I Transparency Be Articulate Least Effort Thus a gd39 b is acceptable in C if a d and gd39 b is not acceptable in C ie if for each 0 of the same type as d and for each acceptable sentence completion b C adandc b ltgtac b 150 A Toy Syntax I Syntax Generalized Quanti ers Q Q Predicates P Pi EiPk Propositi0ns p pi pipk F0rmulas F p not F F and F F or F if F F Qi P P 151 1313 Transparency for all syntactically acceptable b c C p and c b Q c b Claim Transparencyis satis ed Q C p lt IfC p for any c p and c and c have the same contextual meaning hence the result gt Take b to be empty and take c to be a tautology Then Transparency requires that C p and c Q c hence C p and c hence C p 152 not 131339 Transparency for all syntactically acceptable b c C not p and c b Q not c b Claim Transparencyis satis ed Q C p lt IfC p for any c p and c and c have the same contextual meaning hence the result gt Take b to be and take c to be a tautology Then Transparency requires that C not p and c Q not c hence C p and c Q c hence C p as on the previous slide 153 p and gq39 I John is an idiot and he knows that he is incompetent Prediction C John is an idiot 2 John is incompetent Transparency for all syntactically acceptable b c C p and q and c b Q p and c b Claim Transparencyis satis ed Q C p 2 q lt Straightforward once one observes that b must be b can only be because q and c and c are both consti tuents and the only way constituents without can be made parts of larger constituents is by being immediately preceded by not the case here or immediately followed by gt Taking b and c to be some tautology we have C p and q and c Q p and c hence C I p and q Q p hence in particular C p gt q 154 p or gq39 I John is not an idiot or he knows that he is incompetent Prediction C John is an idiot 2 John is incompetent Transparency for all syntactically acceptable b c C p or q and c b Q p or c b Claim Transparencyis satis ed Q C not p 2 q lt Straightforward because 1 orF Q 1 0r notp and F gt Taking b and c to be some tautology we have C p or q and c Q p or c hence C p or q or in other words C I not 19 gt q 155 if p gq39 I If John is an idiot he knows that he is incompetent Prediction C John is an idiot 2 John is incompetent Transparency for all syntactically acceptable b c C ifp qandc b ltgtifp c b Claim Transparencyis satis ed cgt C p 2 q We treat conditionals as material implications lt Straightforward gt Taking b and c to be some tautology we get C ifp q and c ltgt ifp c hence C I if p q 156 General Results I Theorem 1 from AntiDynamics 2007 For a propositional logic with not and or and 13913 this system is fully equivalent to Heim 1983 supplemented with the disjunction of Beaver 2001 not pp presupposes p p and qq presupposes p 2 q p or qq presupposes not p 2 q if pp q presupposes p if p qq presupposes p 2 q but the result applies in full generality not just to unembedded sentences 157 Every P QQ I Each of these 10 students has stopped smoking Prediction C Each of these 10 students smoked Transparency for all syntactically acceptable b c C Every P Q and c b Q Every P c b Claim Transparencyis satis ed Q C Each P Q lt Straightforward gt Taking b and c to be a tautologous predicate we get C Every P Q and c Q Every P c hence C Every P Q 158 N0 P QQ I No of these 10 students has stopped smoking Prediction C Each of these 10 students smoked Transparency for all syntactically acceptable b c C No P Q and c b Q No P c b Claim Transparencyis satis ed Q C Each P Q lt Straightforward gt Suppose for contradiction that i is a Pindividual which is not a QindiVidual Taking b and c to be true of i and nothing else we get C No P Q and c Q No P c The righthand side is false but the lefthand side is true 159 General Results Theorem 2 from AntiDynamics 2007 Under Conditions C1 and C2 the equivalence can be extended to a system that includes any generalized quantifier that satis es Permutation Invariance Extension and Conservativity C1 NonTriviality any quanti cational clause should have a chance of a making a nontrivial contribution C2 The domain has constant size and each restrictors is true of a constant number of individuals throughout C Additional Result This system derives the projective behavior of connectives from their truthconditional contribution and their syntax and hence it is predictive 160 Unless I Unless John didn39t come Mary will know that he is here I a Prediction of Heim 1983 No prediction unless is not discussed b Prediction of Transparency There should be no presupposition if John carne gt John is here This follows from the equivalence Unless John didn39t come q ltgt Unless John didn39t come John came and q 161 While While John worked for the KGB Mary knew that he wasn39t entirely truthful about his professional situation a Prediction of Heim 1983 No prediction While is not discussed b Prediction of Transparency Given knowledge that a spy is not entirely truthful about his professional situation there should be no presupposition This follows from the equivalence While John worked for the KGB q ltgt While John worked for the KGB he worked for the KGB and q 162 Problems a If John is an idiot he knows that he is incompetent b John knows that he is incompetent if he is an idiot a This house has no bathroom or the bathroom is well hidden after Partee b The bathroom is well hidden or this house has no bathroom a If this house has a bathroom the bathroom is well hidden b If the bathroom is not hidden this house has no bathroom Notes pr q z Ifnot q not p Ifnot p and q not p z pr p and q z Ifnot q not p163
Are you sure you want to buy this material for
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'