On Sense and Reference

Gottlob Frege
 The Identity Dilemma

• Either identity is a relation between names or a relation between objects.

– If it’s a relation between names then the truth of identity statements is

a matter of linguistic convention.

– If it’s a relation between objects then all true identity statements are
trivial.

• In either case, we can’t explain how identity statements can be

informative.

– If it’s a matter of linguistic convention it’s arbitrary: we can adopt

whatever conventions we please

– If it’s a relation between “ objects” it just boils down to the relation

between an object and itself, which is obvious and uncontroversial.
(King George didn’t wonder whether Scott = Scott!)
 Compositionality Thesis
• Compositionality: The meaning of a whole sentence is determined by
the meaning of its parts.
– Substitutivity Principle: replacing parts of a sentence with other
expressions that mean the same thing should leave the meaning of
the whole sentence unchanged.
• Why should we care about compositionality?

• Because we want certain characteristics of sentences to be a function of

characteristics of their parts, for example, we want the truth values of
sentences in propositional logic to be a function of the truth values of
their parts.
• Putative counterexamples: failures of substitutivity

– Frege’s Identity Puzzle

– Frege’s Propositional Attitude puzzle

 Frege’ s Identity Puzzle
• In general, sentences of the form a = a are a priori while sentences of the
form a = b are a posteriori

– a priori: can be known “ prior to” experience

– a posteriori (empirical): can only be known on the basis of

experience (i.e. observation, experiment, etc.)

• Example: true, informative identity statements like

– (1) The Morning Star = The Evening Star

• are different in “ cognitive value” from trivially true ones like

– (2) The Morning Star = The Morning Star

• (2) appears to be true in virtue of language alone--everything is identical

with itself--but (1) says something about the world: it was an empirical
discovery.
 Frege’ s Propositional Attitude Puzzle
• Propositional attitudes: ways in which people are related to propositions, e.g.
believing, hoping, fearing, etc.

• We describe people as having propositional attitudes in sentences with

subordinate clauses

– Example: sentences of the form ‘x believes that P’ where ‘P’ stands for
a subordinate clause (P is a complete sentence but ‘x believes that’ isn’t)

• Problem: substitutivity fails within such subordinate clauses, for example

– (3) George believes that Mark Twain wrote Huckleberry Finn.

– (4) George believes that Samuel Clemens wrote Huckleberry Finn.

• (3) may be true even if (4) is false, even though…

 Mark Twain = Samuel Clemens

Mark Twain Samuel Clemens

 Counterexample?

• (3) and (4) don’t have the same cognitive value: George may believe that
Mark Twain wrote Huckleberry Finn but not believe that Samuel Clemens
wrote Huckleberry Finn

– (3) George believes that Mark Twain wrote Huckleberry Finn.

– (4) George believes that Samuel Clemens wrote Huckleberry Finn.

• So it could be that (3) is true but (4) is false

• But (4) is just the result of substituting another name of the same person
for “ Mark Twain” since Mark Twain = Samuel Clemens

• Again, substitutivity seems to be violated

 Toward a solution: the sense/reference distinction

It is natural, now, to think of there being connected with a sign…besides

that to which the sign refers, which may be called the reference of the sign,
also what I should like to call the sense of the sign wherein the mode of
presentation is contained…The reference of ‘evening star’ would be the
same as that of ‘morning star,’ but not the sense

• “ meaning” is ambiguous

• Sense: dictionary meaning, the “ thought” behind an expression or

sentence

– NB: by “ thought” Frege doesn’t mean an idea in someone’s head

• Reference: “ aboutness,” what an expression picks out

 Senses aren’ t ideas

I compare the Moon itself to the reference…mediated by the real image

projected by the object glass in the interior of the telescope, and the retinal
image of the observer.
 Sense
The reference and sense of a sign are to be distinguished from the
associated idea…This constitutes an essential distinction between the idea
and the sign’s sense, which may be common property of many and
therefore not a part of a mode of the individual mind…[different people]
are not prevented from grasping the same sense; but they cannot have the
same idea.

– The sense of a name is an individual concept

– The sense of a predicate is a property or relation

– The sense of a sentence is a proposition

• What are these things???

 Sense and Reference
Sense
• The reference of a name is an individual

A proper name (word, sign, sign combination,

expression) expresses its sense, stands for or
designates its reference.
Ducati

• The reference of a sentence is its truth value

Ducati is a chocolate lab.

The True
 Compositionality Thesis Revised

• The meaning of a whole sentence is determined by the meaning of its parts,

but “meaning” is ambiguous.

• So we need two Substitutivity Principles:

– replacing parts of a sentence with other expressions that have the same
sense leaves the sense of the whole sentence unchanged.

– replacing parts of a sentence with other expressions that have the same
reference leaves the reference of the whole sentence unchanged.

• (An interesting question: why The True or The False as the reference of
sentences?)
 Truth Values as Reference of Sentences

• The thought loses value for us as soon as we recognize that the reference
of one of its parts is missing. We are therefore justified in not being
satisfied with the sense of a sentence, and in inquiring also as to its
reference…Because, and to the extent that, we are concerned with truth
value…We are therefore driven into accepting the truth value of a
sentence as constituting its reference.

• A point to ponder: Are we driven to this?

– Would anything else do as the reference of sentences?

– Why does Frege want truth values (The True or The False) as the
references of sentences?
 Truth Value as the Reference of Sentences

If our supposition that the reference of a sentence is its truth value is

correct, the latter must remain unchanged when a part of the sentence is
replaced by an expression having the same reference. And this is in fact
the case…What else but the truth value could be found that belongs quite
generally to every sentence if the reference of its components is relevant,
and remains unchanged by substitutions of the kind in question?

• Where expressions that have the same reference are substituted in

sentences, truth value remains the same (even if other stuff changes)

– Try it out on the Morning Star/Evening Star example!

• Where substituting expressions in sentences changes truth value we can

conclude that these expressions don’t have the same reference.

– The point of the discussion of the Propositional Attitude Puzzle:

substituting expressions that otherwise have same reference changes
truth value so they don’t have their customary reference.
 Frege’ s Identity Puzzle

• Informative identity statements like

– (1) The Morning Star = The Evening Star

• Are different in “ cognitive value” from trivial ones like

– (2) The Morning Star = The Morning Star

• (2) is an a priori truth of logic since everything is identical with itself--

but (1) says something about the world.

• If identity is a relation between objects then (1) should be as trivial as

(2) and wouldn’t be an important astronomical discovery.

• If identity is a relation between names then (1) is a matter for decision

rather than discovery: we make it true by adopting a linguistic
convention.
 Leaping between the horns

(1) The Morning Star = The Evening Star

(2) The Morning Star = The Morning Star

• Frege needs to explain why some true identity statements, like (2) are
trivial but others, like (1) aren’t.

• He’ll do this by arguing that in true, non-trivial identity statements, the

expressions on either side of the identity have the same reference but
different senses.

• He’ll go on to use that distinction in developing a theory that saves

compositionality from putative counterexamples—like the Morning
Star/Evening Star problem.
 A Misguided Attempt

Maybe “The Morning Star” and “The Evening Star” don’t

mean the same thing.

M.S. E.S.

4 am 8 pm

Maybe The Morning Star consists of those temporal parts of Venus we

see during mornings while The Evening Star is the sum of temporal
parts of Venus we see during evenings.
 Why this solution fails

(1) The Morning Star = The Evening Star

(2) The Morning Star = The Morning Star

• Different temporal parts of the same thing are not identical!

• If “ The Morning Star” and “ The Evening Star” name different temporal
parts of Venus then (1) is FALSE! The Morning Star ≠ The Evening Star!

• Similarly, Samuel Clemens took the pen name “ Mark Twain” as an adult
but “ Mark Twain” does not just name his adult person-stages: Mark
Twain was born in Hannibal, MO even though as a baby he wasn’t called
“ Mark Twain.”
 Some facts about identity

• Identity is an equivalence relation

– Reflexive: x = x

– Symmetric: if x = y then y = x
An equivalence relation partitions a
– Transitive: if x = y and y = z then x = z set into equivalence classes.

• Identity is an indiscernibility relation

– If x = y, then whatever property x has y has and vice versa

– Problem: this principle is sometimes stated as “ x = y iff whatever is

true of x is true of y and vice versa” but not every grammatical
predicate assigns a property to the object of which it is true!
 Indiscernibility of Identicals

• If x = y then x and y have exactly the same properties

– Example: The Morning Star = The Evening Star therefore since the
Morning Star is really a planet, The Evening Star is really a planet.

– Example: Mark Twain = Samuel Clemens therefore since Mark

Twain has the properties of being a journalist, being the author of
Huckelberry Finn, etc. Samuel Clemens has those properties.

• So if two names, a and b, refer to different temporal (or spatial) parts of

an object, a ≠ b!
 Solution to the Identity Puzzle

(1) The Morning Star = The Evening Star

(2) The Morning Star = The Morning Star

• What makes (1) and (2) true is that fact that the referents of “ The Morning
Star” and “ The Evening Star” are the same.

• The same fact about the world makes them true

• But the sense of these expressions isn’t the same

 Same sense, same reference

(1) The Morning Star = The Evening Star

(2) The Morning Star = The Morning Star

• Where the senses of expressions are the same, they refer to the same
thing.

• So we know the expressions on either side of the identity in (2), i.e

“ The Morning Star,” refer to the same thing--hence that (2) is true.

• Synonomy is sameness of sense--so we know also that, e.g. the tallest

bachelor in the class is the tallest unmarried male in the class.

• We know these things by language alone

 Different sense, same reference

(1) The Morning Star = The Evening Star

(2) The Morning Star = The Morning Star

• But expressions with different sense may refer to the same thing

• And we can’t determine whether or not they do that just by language alone

• So The Morning Star = The Evening Star was an important astronomical

discovery
 Different sense, same reference

a c
b

The point of intersection of a and b is then the same as the point of intersection of
b and c. So we have different designations for the same point, and these names
('point of intersection of a and b', 'point of intersection of b and c') likewise
indicate the mode of presentation; and hence the statement contains actual
knowledge.
 Substitutivity vindicated!

• (1) and (2) express different thoughts since “ Morning Star” and
“ Evening Star” have different senses so a person can know that (2) is
true without knowing that (1) is true.

• The senses of both (1) and (2) are determined by the senses of their
parts!

– They express different proposition so a person might know that one

is true without knowing that the other is

• The reference of (1) and (2) is determined by the reference of their

parts

– The reference of both sentences is The True

• Our mistake was confusing sameness of reference with sameness of

sense!
 Frege’ s Propositional Attitude Puzzle

(3) George believes that Mark Twain wrote Huckleberry Finn.

(4) George believes that Samuel Clemens wrote Huckleberry Finn.

• “ Mark Twain” and “ Samuel Clemens” have the same reference—like

“ The Morning Star” and “ The Evening Star” but

• But that’s no guarantee that (3) and (4) have the same truth value!

• Substitutivity fails!

• So it looks like the solution to Frege’s identity puzzle doesn’t work for
Frege’s Propositional Attitude puzzle!
 Problem with truth value

• Remember the reference of a sentence is supposed to be its truth

value.

• And the reference of the whole is supposed to be determined of the

reference of its parts.

– (4) George believes that Samuel Clemens wrote Huckleberry Finn

• seems to be the result of substituting an expression that has the same

reference as “ Samuel Clemens,” i.e. “ Mark Twain,” in

– (3) George believes that Mark Twain wrote Huckleberry Finn

• But (3) is true and (4) is false!

 A misguided solution

(5) Mark Twain wrote Huckleberry Finn

(6) Samuel Clemens wrote Huckleberry Finn

• Mark Twain = Samuel Clemens BUT “ Mark Twain” ≠ “ Samuel

Clemens”

• George may believe that sentence (5) makes a true statement but not
that (6) makes a true statement but that’s because his beliefs are about
different sentences.

• This is no worse than recognizing that the English sentence makes a

true statement but not recognizing that the French one does.

– (E) The book is on the table.

– (F) Le livre est sur le table.

 Why this solution fails

• George’s beliefs are not about a sentences but about people and books:
he believes that Mark Twain wrote Huckleberry Finn and that the
book is on the table.

• Seeing a book on the table, he wouldn’t say “ le livre est sur le table”
or believe that that sentence was true but

• a French speaker would be perfectly correct in saying, “ George crois

que le livre est sur le table.”

• But incorrect in saying, “ George crois que la phrase ‘le livre est sur le
table’ est vrai.”
 Sentences and Propositions

• George believes the book is on the table.

• George believes the proposition that speakers of other languages

would express as, e.g. “ le livre est sur le table,” “ il libro é sul tavolo,”
etc.

– When English, French and Italian speakers say these things

they’re uttering the different sentences but expressing the same
proposition.

• But he does not believe that those non-English sentences express true
propositions because he does not know what the heck those sentences
mean.
 So we still have a problem…

• George believes that Mark Twain wrote Huckleberry Finn.

• Someone in the know would affirm that George believes it was Samuel
Clemens that wrote the book and not someone else

– Compare George who believes that that guy wrote the book even
though he doesn’t know that guy was also called Samuel Clemens

– With someone who believes that Huckleberry Finn was really written
by H. L. Mencken.

• But George does not believe that Samuel Clemens wrote Huckleberry
Finn.
 Substitutivity Fails

(7) George believes that ___ wrote Huckleberry Finn

• (7) (apparently) designates the property of being-believed-by-George-to-be-

the-author-of-Huckleberry-Finn

– Plugging a name into the blank (apparently) ascribes that property to the
individual to which the name refers, e.g. plugging in “ Mark Twain”
ascribes that property to Mark Twain—a.k.a. Samuel Clemens.

• George has heard the author referred to as “ Mark Twain” ; George has never
heard the name “ Samuel Clemens” in that connection

• So filling the blank with “ Mark Twain” gets a true sentence while plugging in
“ Samuel Clemens” doesn’t.
 How can this be?!!?!

• Either a thing has a property or it doesn’t

– The name by which we call it or the way in which we describe it

shouldn’t make any difference to whether it has the property or not.

– So how come the (supposed) property of being-believed-by-George-

to-be-the-author-of-Huckleberry-Finn attaches to this guy when we
call him “ Mark Twain” but not when we call him “ Samuel
Clemens” ???

• Subordinate clauses like “ Mark Twain wrote Huckleberry Finn” as it

occurs in (3) pose a problem because we want the truth value of a
compound sentence to be a function of their sentential parts.

• The problem is that “ x believes that…” isn’t truth functional!

 Short answer

(7) George believes that ___ wrote Huckleberry Finn

• (7) doesn’t really ascribe a property to the guy whose name plugs in the
blank

• In general, not all contexts in which names occur ascribe properties to the
referents of those names

• Why this is so, is a long story!

 Propositional attitudes
• “ believe…” , and other propositional attitude verbs that introduce
relative clauses create contexts where co-referential expressions
cannot be intersubstituted salve veritate.

(3) George believes that Mark Twain wrote Huckleberry Finn.

(4) George believes that Samuel Clemens wrote Huckleberry Finn.

• In (3) and (4) what George believes are the propositions expressed by
(5) and (6), i.e. their senses.

(5) Mark Twain wrote Huckleberry Finn.

(6) Samuel Clemens wrote Huckleberry Finn

 Propositional Attitudes: Frege’ s Solution
• In direct discourse expressions have their “ customary” sense and
reference.

– Direct discourse: we’re simply expressing a proposition; we’re not

quoting someone else or talking about someone else’s believing a
proposition.

• So the reference of “ Mark Twain” and “ Samuel Clemens” in the

following sentences is the same--a certain person who is the
“ customary” reference of that name.

– (5) Mark Twain wrote Huckleberry Finn.

– (6) Samuel Clemens wrote Huckleberry Finn.

In (5) and (6), “Mark

Twain” and “Samuel
Clemens” refer to this guy
 In indirect discourse reference shifts

Individual property
sense concept proposition

Mark Twain wrote Huckleberry Finn.

reference The True

George believes that Mark Twain wrote Huckleberry Finn.

reference proposition
 Senses of Names and Sentences
• The sense of “ Mark Twain” is something like “ the famous journalist
and author who wrote Huckleberry Finn, etc.”

• The sense of “ Samuel Clemens” is something like “ Mr. and Mrs.

Clemens’ kid Sam”

• So the customary senses of “ Mark Twain” and “ Samuel Clemens” are

different.

• Since the sense of the whole is determined by the sense of its parts,(5)
and (6) have different senses, i.e. express different propositions.

(5) Mark Twain wrote Huckleberry Finn.

(6) Samuel Clemens wrote Huckleberry Finn

 Indirect Discourse: Reference shifts

(3) George believes that Mark Twain wrote Huckleberry Finn.

(4) George believes that Samuel Clemens wrote Huckleberry Finn.
(5) Mark Twain wrote Huckleberry Finn.
(6) Samuel Clemens wrote Huckleberry Finn

• In indirect discourse and propositional attitude ascriptions, e.g. (3) and (4),
the customary senses of expressions become their references.

• So in (3) the reference of “ Mark Twain wrote Huckleberry Finn” is the

sense of (5).

• In (4) the reference of “ Samuel Clemens wrote Huckleberry Finn” is the

sense of (6).
 Where we went wrong

(3) George believes that Mark Twain wrote Huckleberry Finn.

(4) George believes that Samuel Clemens wrote Huckleberry Finn.

• Our mistake was thinking that the embedded sentences in (3) and (4)
had their customary reference, i.e. The True.

• If that were so then substituting one for the other shouldn’t change the
truth value of the whole sentences.

• But the reference of the embedded sentences has shifted to their

customary senses, which are different!
 So reference of the sentences may be different

(3) George believes that Mark Twain wrote Huckleberry Finn.

(4) George believes that Samuel Clemens wrote Huckleberry Finn.

• The reference of the parts of (3) and (4) are different so the
references of the whole sentences (3) and (4) can be different.

– The embedded sentences refer to the customary senses of (5) and

(6) respectively

– And these senses are different!

• So there is no reason why the reference, i.e. truth value of (3) and (4)
should be the same!
 Some predicates attribute properties to objects

• Example: “ ___ is red”

• The name that fills the blank refers to an individual

• The predicate “ ___ is red” designates a property

• The complete sentence, when a name is filled in, says that the
individual to which the name refers has that property.
 Some predicates don’ t assign properties to objects

• Example: “ George believes that ____ wrote Huckleberry Finn.

• The name that fills this blank doesn’t refer to an individual

• The predicate doesn’t designate a property

• The complete sentence doesn’t say that the individual whose name fills
the blank has the property of being-believed-by-George-to-have-
written-Huckleberry-Finn
 Frege’ s analysis: summary

• “ George believes that ___” is about the relation between George and a
proposition, viz. the sense of the sentence that fills the blank.

• It’s not about a relation between George and Mark Twain, a.k.a.
Samuel Clemens.

• “ Mark Twain wrote Huckleberry Finn” and “ Samuel Clemens wrote

Hucklberry Finn” express different propositions

• So you can’t substitute one for the other salve veritate in indirect
discourse such as “ George believes that ____”
 Frege’ s theory solves both puzzles

• Benefits: saves Indiscernibillity of Identicals, explains restrictions on

substitutivity, explains how true identities can be informative.

• Costs: commits us to:

– individual concepts

– properties

– propositions

Creatures of
Darkness!

W. V. O. Quine
Can we avoid the netherworld of abstracta?

or is this..

The End