Talk:Sloppy identity

== Planned changes as part of course project (Ling300 2017wT1) We will be making changes to this entry as part of a course project. — Preceding unsigned comment added by RM Dechaine (talk • contribs) 21:08, 24 November 2017 (UTC)

Intro
In linguistics, sloppy identity refers to cases where an ellipsis gives rise to a meaning which differs along some dimension from the strict meaning of its antecedent, thereby leaving two possible interpretations of a given sentence.

Consider the following basic example of sloppy identity:

(1) John scratched his arm and Mary did too.

This sentence can have a “strict” interpretation:

i) Johni scratched hisi arm and Maryj did too. (scratch hisi arm)

Or a “sloppy” interpretation:

ii) Johni scratched hisi arm and Maryj did too. (scratch herj arm)

As shown by the above examples, in the strict reading both the elided pronoun and the pronoun in the antecedent is referring to John and therefore share the same index. On the other hand, the elided pronoun in the sloppy reading is referring to Mary, and therefore does not share the same index as the pronoun in the antecedent.

Two pronoun ellipsis puzzle:
(first noted by Dahl, 1973)

(2) John said he saw his mother and Bill did too. (said he saw his mother)

Four possible readings:

i) Bill said John saw John’s mother

ii) Bill said Bill saw Bill’s mother

iii) Bill said Bill saw John’s mother

iv) *Bill said John saw Bill’s mother

The 4th reading is said to be “degraded or impossible” due to the c-command generalization : A strict pronoun may not c-command a sloppy pronoun.

=== Two pronoun attitude puzzle : === (first noted - Percus and Sauerland, 2003)

(3) John dreamed that he was marrying his grand-daughter

Four possible readings:

i) John dreamed that John was marrying John’s grand-daughter

ii) John dreamed that Bill was marrying Bill’s grand-daughter

iii) John dreamed that Bill was marrying John’s grand-daughter

iv) *John dreamed that John was marrying Bill’s grand-daughter

Sloppy read without elision :
(4) Everyone loves someone

Two possible readings:

i) For every person in the world there is some other person who they love

ii) There is one person in the world that everyone else in the world loves

History
The discussion of “sloppy identity” amongst linguists dates back to a paper by John Robert Ross in 1967, in which Ross identified the interpretational ambiguity in the elided Verb Phrase (from now on VP) of the previously stated sentence (1) found in the intro.

Ross tried, and failed, to properly account for instances of “sloppy identity” using syntactical structure. It was not until the advent of “Derived VP” rules, first introduced by linguist Lawrence Bouton (1970), that syntacticians were able to correctly explain issues of “sloppy identity”. Shortly there after, using a new found extension of Bouton's Derived VP rule, Sag proposed the deletion (Derived VP) approach. Ross proposed the definition: “constituents are identical if they have the same constituent structure and if they are identical morpheme-for-morpheme, or if they differ only as to pronouns, where the pronouns in each of the identical constituents are commanded by antecedents in the non-identical portions of the phrase-marker.” He noted his theory predicts too many ambiguities. Consider the sentence: (5) John told Bill that he was smart, and Sam told Harry.

According to Ross’ theory, the pronoun “he” would be commanded by its antecedent “John”. In the second constituent “Sam told Harry that he was smart”, the pronoun “he” would be commanded by the antecedent in a different position of the phrase marker, namely “Harry”. So the reading of this example is: John told Bill that John was smart, and Sam told Harry that Harry was smart. The problem with reading under Ross’ theory is that it does not allow non-referential readings, where we consider two constituents identical in form do not refer to the same subject.

Lawrence Bouton (1970) was the first to develop a syntactic explanation of VP-Deletion, commonly referred to by contemporary linguists as VP - ellipsis. Boutons theory of VP-Deletion and Ross’ observation of sloppy identity served as an important foundation for linguists to build on.

In Bouton’s explanation of VP deletion we can understand that in order for the VP to be deleted, there must be a syntactically identical VP in the sentence that c-commands the deletable VP.

Theories
Building on the work of Bouton (1970) and Ross (1969), Barbara Partee (1975) developed what has come to be one of the most important and influential approaches to explain VP to date, the Derived VP-Rule, which introduces a null operator at the VP level. Shortly thereafter, Ivan A. Sag (1976) developed the Deletion Derived VP approach, and Edwin S. Williams (1977) developed the Interpretive Derived VP Approach. These rules are still used by many today.

Derived VP-Rule:
(Partee 1975) This introduces a null operator at the VP level and is a very important rule that has been built on over the years to make sense of sloppy identity. This VP-rule is used by many linguists as a foundation for their ellipsis rules.

Deletion Derived VP approach:
(Sag 1976)

Ivan Sag proposed two logical forms, one of which the coreferential pronoun is replace by a bound variable. This leads to the rule of semantic interpretation that takes pronouns and changes them into bound variables. Abbreviated   Pro ->BV

A simple sentence like
(6) [Betsyi loves heri dog]

where the non-sloppy reading being "Betsy loves her own dog" can undergo Pro->BV and we could get

(6.a) [Betsyi loves herj(or x’s) dog]

Where herj is someone else or x’s is anyone else's dog

A more complex example
(7) Betsyi loves heri dog and Sandyj does ∅ too where ∅ = loves her dog

The non-sloppy reading of (7) is " Sandy loves Betsy’s dog as well" so the ∅ is a VP that has been deleted and we can see why with the explanation that follows

(7.a) Betsyi λx(x loves heri dog) and Sandy λy(y loves heri dog

So, the VPs that are being represented by λx and λy are syntactically identical and therefore the one that is being c-commanded (λy) can be deleted. This then forms

(7.b) Betsyi loves heri dog and Sandyj does ∅ too

The sloppy reading is saying Sandy loves her own dog, which can be represented in the following form

(7.b.i) Betsyi λx(x loves heri dog) and Sandyj λy(y loves herj dog

Since the embedded clauses are identical, the logic of this form is that the variable x must be bound to the same noun phrase in both cases. Therefore “Betsy” is in the commanding position that determines the interpretation of the second clause.

The Pro → BV rule that converts pronouns into bound variables can be applied to all the pronouns. This then allows for

(6.a) “Betsyi λy(y loves heri dog)” to

(6.b) “Betsyi λy(y loves y’s dog)” and he whole sentence in (7) then becomes

(7.c) Betsyi λx(x loves x’s dog) & Sandyj λy(y loves y’s dog)

Another way the VP can be syntactically identical is, if λx(A) and λy(B) where every instance of x in A has a corresponding instance of y in B. So, like in the example above, for all instances of x there is a corresponding instance of y and therefore they are identical and again the VP that is being c-commanded can be deleted.

Other Explanations of Sloppy Identity
There are subtler ambiguities that exist which cannot be explained in syntactic terms of deletion and substitution rules.

Take these example:

(8) John lives in New York, and Bill lives in the United States, too.

This sentence has the form of two identical constituents, but it is not an environment where sloppy identity would occur because it bears a causal relationship and presupposes the listeners know New York is in the United States.

(9) John thinks he is smart, and Bill suffers from the same delusion.

There are sentences that bear ambiguity even though it does not have the VP deletion transformational processes that occur in structures with identical constituents.

(10) Only Sam loves his wife.

Possible interpretations can be 1) Sam is the only person such that that person loves his wife, and 2) Sam is the only person that loves Sam’s wife. This ambiguity is not explained by the above mentioned syntactic theories of VP ellipsis because there are no identical constituents to be deleted.

(11) Peter has got a letter from his mother, too.

The use of too “presupposes that the property that the predicate of the sentence attributes to the subject also holds true about someone or something else.” But in this example, the presupposition is not expressed in the words, so it allows two readings:

i) where someone else has got a letter from his own mother

ii) where someone else has got a letter from Peter’s mother.

(12) Even John distrusts his wife.

This allows two readings:

i) other people involved in this context distrust their own wives

ii) other people involved in this context distrust John’s wife.

These readings depend on whether there is an implicit presupposition that someone else distrusts John’s wife too.

Dcstrosh (talk) 07:25, 5 November 2013 (UTC)

Horribly inaccessible
This article is horribly inaccessible. It is terribly spread out, filled with acronyms (PF, LF, BV, etc.), and the information is largely particular to one particular approach to syntax and semantics. The use of punctuation and abbreviations is inconsistent. The information here is accessible to only the narrowest of audiences. --Tjo3ya (talk) 06:00, 7 December 2017 (UTC)

UBC LING300 WikiProject (2017wT1)
As a course project for the LING300 course at UBC, we were given the topic Sloppy Identity, and we are planning to expand on the section Sloppy Identity in Mandarin. In terms of content, we will include shi-support and Mandarin sluicing. For shi-support, we will give examples and explanations. We will also include how negation and questions work in terms of shi-support. For Mandarin sluicing, we will discuss the following three properties: c-commanding, lexical entity between wh-words, and the "na" effect, giving explanations and examples for each. In terms of editing, we will attempt to make the article more accessible by making the style more consistent, expanding abbreviations, and linking the relevant page for readers to access if they wish to do so. -- SCHU ( talk) 14:49, 7 December 2017 (UTC)