GRICE AND URMSON

Speranza

Urmson, like Grice, wants to examine a group of expressions (Grice uses "E" for expression) which are not usually considered as a group.

Many of these verbs, including such important ones as "know", "believe", and "deduce" are frequently misconstrued by philosophers, and their consideration as a group may help to get them into better perspective.

None of these verbs is here examined exhaustively.

In general only that which can be said of all is said of each.

For convenience, this group Urmson calls the group of "parenthetical" verbs.

No great significance should be attached to this title.

Such significance as it has can more conveniently be explained later.

Urmson then proposes a delimitation of the group of parenthetical verbs.

In prose, the verb "to read" is used in the present continuous form:

"I am reading"

to report a contemporary happening.

-- cfr.

Grice, "I am hearing a noise" ("Personal Identity", Mind, 1941).

The present perfect form

"I read."

 is used only to report what one often, or habitually, does.

This is true of most of the verbs in the English language (and other languages as well such as Greek or Latin).

It has been observed in recent years, but only, even by philologists, in recent years, that some verbs
do NOT conform to this pattern.

Some verbs either have no present continuous tense, or, when they have, it is only so in one out of two easily distinguishable uses.

Thus the verb, "to prefer", has no present continuous tense.

We never do say

"I am preferring" -- cfr. Macdonald, "I'm lovin' it".

The verb "to wish" can have a present continuous form, as in

"I wish whenever I pass a wishing-well"

"I am wishing at a wishing-well."

-- buut has not when we say

"I wish that you would make up your mind."

Here, in the third example, the use of "wish" is similar to the use neither of "I wish" (simpliciter) nor of  "I am wishing" in the first two examples.

It is clear, then, that these verbs are not simply defective of a present continuous tense, having only a normal present.

For "I prefer" and "I wish" (as used in the third example) are NOT used in the way that 'I read" and "I play" are used.

Some of these anomalous verbs are used normally with a direct object.

Examples are "love", "like", "prefer".

Some are used normally with a subjunctive or other non-indicative clause.

Examples are "wish", "command", " beg", "beseech".

These verbs do NOT fall within the group of parenthetical verbs, and are not further discussed by Grice and Urmson (on that occasion), though they are well
worth discussing as groups.

Grice and Urmson intend to discuss only a special set of verbs which lack a present continuous tense s which must now be distinguished from the others.

Let us start with an example.

Taking the verb

"to suppose"

we may note that In the first person present we can idiomatically say any of the following :

"I suppose that your house is very old." -- INITIAL POSITION

"Your house is, I suppose, very old." ----- MIDDLE POSITION

"Your house is very old, I suppose." ------ FINAL POSITION

A verb which, in the first person present, can be used,, as in the example above, followed by 'that* and an indicative clause, or else can be inserted at the middle or end of the indicative sentence, is a parenthetical verb.

Note that this is a grammatical distinction, and that these verbs are called
parenthetical because of this grammatical feature of their use.

("Parenthetical" Is sometimes used of a piece of information slipped into another context, but Grice and Urmson do NOT wish to imply that these verbs are parenthetical in any sense except that they
are sometimes used parenthetically in a purely grammatical sense.

Beyond that "parenthetical" is merely a convenient
label.

In some contexts it will be virtually indifferent, on all
but stylistic grounds, whether the verb occurs at the beginning, middle, or end of the indicative sentence with which it is conjoined.

This will not always be so, but when it is the verb
will be said to be used purely parenthetically.

Thus in most
contexts

"l suppose that your house is very old."

 would be used purely parenthetically, for it would mean virtually the
same as either

"Your house is very old, I suppose."

or

"Your house is, I suppose, very old." -- cfr. Speranza's favourite preferred option: "Your house, I suppose, is very old" -- and

"Your house is very, I suppose, old."

--

"Your, I suppose, "house" is very old."

"Your house is very, I suppose, old."

---

If one person says

"I suppose THAT your house is quite new"

 and another says

"Well, I suppose THAT it is very old",

 in the latter statement the verb "to suppose" is NOT being used purely parenthetically.

Grice and Urmson study parenthetical verbs in their more or less pure parenthetical use for the sake of simplicity.

On other occasions most of what Grice and Urmson have to say will remain true, but
will be more or less far from being the full story.

It would
be perhaps more accurate to say that the features of parenthetical verbs to which Grice and Urmson draw attention are one aspect
of their use which is relatively more important on the occasions
on which Grice and Urmson concentrate than on others.

But it is convenient to talk of a parenthetical use.

A purist (like Speranza) may substitute "aspect" for use throughout -- and Speranza is following his tutor (who quotes from a Greek grammar).

Another preliminary point must be made before we get down to philosophical business.

Part of what Grice and Urmson design to
show is how differently these verbs are used in the first person
present and in other persons and tenses.

Therefore Grice and Urmson at first confine their attention to their pure parenthetical use
in the first person present.

It will be no accident therefore
that all examples will be in this person and tense, nor will
it be an objection to Grice's and Urmson's thesis that what I say will not be true
if examples in other persons or tenses are substituted for the
ones given.

It  will in fact be a partial confirmation of Grice's and Urmson's
thesis.

A random and Incomplete list of parenthetical, verbs might
be helpful at this stage.

know
believe
deduce
rejoice,
regret
conclude
suppose
guess
expect
admit
predict.

A
few minutes' reflection will enable anyone to treble this list

Some of these verbs, like "conclude", are ALWAYS parenthetical,
though of course not always used purely parenthetically.

Others, like rejoice, may be non-parenthetical and have a
present continuous tense.

Grice and Urmson are  only concerned with
these verbs when they are parenthetical.

We shall find easy
tests to distinguish their different uses.

Parenthetical verbs are not psychological descriptions.

Let
us take for comparison three sentences ;


"I rejoice whenever my sailor brother comes home."

"I am rejoicing because my sailor brother is home."

"I rejoice that you have returned home at last."

In the first two utterances, "rejoice" is not a parenthetical verb.

In the first utterance, the main verb reports the periodic recurrence of a
psychological condition, the occasions of which are given In
the subordinate clause.

In the second utterance the main verb reports that
something is going on now, and the subordinate clause states
the cause.

No such explanation can be given of the third utterance, where
rejoice is a parenthetical verb though not used purely parenthetically.

The point becomes even clearer If we contrast a
purely parenthetical use with a clearly descriptive verb :

"He is, I regret, unwell."

"I am miserable because he is unwell."

Note that it would be ABSURD to say :

"He is, I am miserable, unwell." (But cfr. "He is, and I'm miserable for it, unwell.")

"I regret because he is unwell." (but cfr. "I regret his being unwell").

'I am miserable' does, *I regret* does not, describe a psychological condition.

In

"I am misearble because he is unwell"

 "because he is unwell"

gives the
cause of a mental state.

------------------------------------------
"I regret because he is unwell" is absurd (rather than merely otiose)

because a cause is given where nothing has been

 described needing a causal explanation.
---------------------------------------------------

It should surely be obvious that though we are, in some sense, dealing with psychological or mental
verbs, they are not parts of psychological histories as are
verbs like "ponder" or "be miserable".

Nor, so far as Grice and Urmson can see,
is it any more plausible to say that they report dispositions
to behave in certain ways.


--- cfr. Grice, "Intention and disposition", 1944 -- The Grice Papers.



This has, however, not seemed
obvious in the case of some parenthetical verbs to some philosophers.

For while the difficulty of regarding

"I know"

and

"I believe"

as If they reported contemporary events -- as if we
said

"I am knowing that p"

 and

"I am believing that p" -- has been appreciated, many philosophers tend to treat know and believe as
though they were simply defective of a present continuous
tense.

Thus 'I know' and *I believe' have been construed
as ordinary present perfects implying, not the frequent truth
of * I am knowing' and "I am believing" but of * I am doing
this thing and the other thing'.

It is an alternative to this
mistaken view that will shortly be given.

Implied claims to truth.

Whenever anyone utters a sentence which could be used to convey truth or falsehood
there is an implied claim to truth by that person, unless the
situation shows that this is not so (he is acting or reciting or incredulously echoing the remark of another).

This needs an
explanation.

Suppose that someone utters the sentence

"It will rain tomorrow" (or "It rained yesterday", or "It's raining").

 in ordinary circumstances.

This act
carries with it the claim that it is true that it will rain tomorrow.

By this is meant that just as it is understood that
no one will give orders unless he is entitled to give orders, so
it is understood that no one will utter a sentence of a kind
which can be used to make a statement unless he is willing
to claim that that statement is true, and hence one would be
acting in a misleading manner if one uttered the sentence if
he was not willing to make that claim.

---

The word 'implies'
is being used in such a way that if there is a convention that
X will only be done in circumstances Y, a man IMPLIES that
situation Y holds if he does X.

---

This point has often been made before, though not always


In these terms, and it is, Urmson and Grice believe, in substance uncontroversial.

Grice and Urmson  wish then to make the point that when a speaker uses a parenthetical verb of himself with an indicative sentence, there is not merely an implied claim that the whole statement is true but also that is true.

This is surely obvious in some cases

"I believe it will rain"

"He is, I regret, too fast."

"You intend, I gather, to refuse."

 But Grice and Urmson think that a little thought shows that it is also true in the case of, say,

"I hear that he is ill in bed"

"He is, I hear, ill in bed"

We
should not and would not say these things if we did not accept
the reports on which our statements were based, and by saying
them WE IMPLY a claim to their truth.

---- Grice is careful here in that the belief is EXPRESSED rather than IMPLICATED, though -- vide, Way of Words, iii -- "Moore's paradox".


The claim to truth need not be very strong.

We shall indeed find that the whole
point of some parenthetical verbs is to modify or to weaken
the claim to truth which would be implied by a assertion.

But even if we say

He is, I suppose, at home.

or

" I guess that the penny will come down heads"

we IMPLY, with however little reason, that this is what we accept as true,

Positive examination of parenthetical verbs.

We
our statements

In contexts* social as well as logical.

For
example, we often have an emotional attitude to the fact we state, or it is likely to arouse emotion in our hearers.

To
some extent, both by accident and by design* our manner,
intonation* and choice of words betray our attitude and prepares our hearers.

But this is imprecise and uncertain, and ?
in writing, is difficult to get right for all but the great stylist.

Again, content and manner give some clue to the hearer or
reader of how he is to understand the statement In relation
to its logical context, but not infallibly.

Further, we make
our statements sometimes with good, sometimes with moderate, sometimes with poor evidence.

Which of these situations we are in need not be obvious to the hearer, and it
would be cumbersome always to say explicitly.

It is Grice's and Urmson's contention that parenthetical verbs are one of the sets of devices
that we use in order to deal with these matters, though not the only set.

By them we prime the hearer to see

the emotional significance,
the logical relevance, and
the reliability of our
statements.

This we do not by telling him how we are moved
or how he should be moved by them, nor by telling him how
our statement fits into the context, nor by describing the evidential situation, but by the use of warning, priming or
orientating signals.

We show rather than state.

This is the
contention which will now be somewhat elaborated,

Suppose that I go to a mother in wartime as a messenger
to Inform her of the death of her son.

I can, no doubt, merely
say

'Madam, your son is dead'.

But this would be abrupt
and harsh, and I would more probably say

Madam, I regret
that your son is dead.

For anyone other than a great actor
it is easier to steer a course between callousness and false
sentiment as a stranger bearing news by the use of a parenthetical verb in this way than by means of intonation.

Clearly I am mainly bearing news, and the addition of "I regret" (not
necessarily at the beginning of the sentence) shows without it being actually SAID that it is being offered, and will be
received as, sad news.

I am not being a hypocrite, even within the excusable, conventional, limits of hypocrisy, if I
personally have no feelings on the matter at all.

Messengers
of that sort can rarely have much feeling in wartime about
each case.

If, for the moment^ we turn to a less purely parenthetical use of the same verity we shall find that the essential
point remains the same.

If, as a friend of the family, I go to
the mother when the death is well known and say

"I much regret that your son is dead. He was a dear friend."

then, no
doubt, I am no longer mainly bearing news.

But I am still
NOT describing my feelings.

It is rather that the signal is
being made for its own sake as an act of sympathy, the indicative clause giving the occasion of my sympathy.

Regret
and rejoice are two of the most obvious examples of verbs
which give emotional orientation when used parenthetically,


Another set of these verbs is used to signal how the statement is to be taken as fitting logically into the discussion.

"I admit that he is able."

assigns the statement that he is able
to the logical position of being support for the opposed
position, or a part of the opposed position which will not be
assailed one shows while saying that he is able that this
is to be treated as an admission.

---- Grice will go on to refer to this as a second-order metaconversational device ("Retrospective Epilogue", Way of Words" --- a non-central speech act ('opposing', 'objecting', 'adding', 'contrasting' -- central spech acts are stating and ordering, only.

One is forestalling a possible
misapprehension.

But don't you see, that is part of my
point.

One is not reporting the occurrence of "a bit of
admitting", whatever that may be supposed to be.

The parenthetical verb in

"Smith was, I conclude, the murderer"

assigns
to the statement the status of following from what has been
said before, preventing it from being taken as, say, an additional fact to be taken into account.


--- Cfr. Grice, "Smith is, therefore, the murderer." -- and the 'conventional implicature' of "therefore".

There is no specific activity of "concluding".

Other verbs which fulfil approximately similar tasks are

deduce
infer
presume
presuppose
confess
concede
maintain
assume.

Another rough group is constituted by such verbs as

know
believe
guess
suppose
suspect,
estimate, and, in a
metaphorical use, feel.

-- never 'sense' -- metaphorical USE -- cfr. Grice's modified Occam's razor -- do not multiply senses beyond necessity, and Urmson's having learned the lesson -- his earlier, "On the two senses of 'probable'".

This group Is probably more controversial than the previous ones, and will require more explanation.

This Is the group which Is used to indicate the
evidential situation In which the statement is made (though
not to describe that situation), and hence to signal
degree of reliability Is claimed for, and should be accorded
to, the statement to which they are conjoined.

Thus

"I think that this Is the right road to take"

Is a way of saying that
this Is the right road, while indicating that one is Just
plumping and has no information, so that the statement will
be received with the right amount of caution ;

"I know", on the other hand,
shows that there is all the evidence one could need ; and so on.

------ The "Scalar" --

Some of these verbs can clearly be arranged in a SCALE showing the reliability of the conjoined statement according to the wealth of evidence:

"know" "believe" "suspect" "guess" .



and adverbs can make the situation even plainer

"I strongly believe"

"I rather suspect"

 and so on. We are,
in fact, In a position where we can either make our statements 'neat', and leave it to the context and the general
probabilities to show to the hearer how much credence he
should give to the statement ; or, In addition to making the
statement we can actually describe the evidential situation
in more or less detail ; or give a warning such as,

"Don't rely
on this too implicitly, but . . .' ; or I can employ the warning
device of a parenthetical verb

"I believe It will rain."

If this
is insufficient for any reason (perhaps It is an Important matter),
then the hearer can ask why and get the description of the
evidential situation.

More will have to be said about these
verbs, but it will be convenient first to introduce another topic, and that is the one of adverbs corresponding to parenthetical verbs.

Grice and Urmson mentioned that parenthetical verbs were not the only device that
we have for warning the hearer how our statements are to be
taken while making it.

It will perhaps make It clearer how
parenthetical verbs are used if one of these other devices Is
briefly outlined.

We were taught at school that an adverb
modifies a verb.

But this is inaccurate, for some adverbs are quite as loosely attached to whole SENTENCES as are parenthetical
verbs.

Examples are :

luckily
happily
unfortunately
consequently
presumably
admittedly
certainly
undoubtedly
probably
possibly
otiosely

and Speranza's favourite

hopefully (Speranza _means_ hope).


Note that the position of these adverbs
is variable In relation to the sentence as In the case of paren-
thetical verbs.

We can say

Unfortunately he Is ill '

or

He is,
unfortunately, ill*.

If the word * modify' Is to be used these
adverbs can perhaps be said to modify the whole statement
to which they are attached.

But how do they modify them ?


Surely by giving a warning how they are to be understood,
Luckily ! , happily and unfortunately indicate the appropriate
attitude to the statement, for example.

Admittedly, consequently and presumably, among others, Indicate how to
take the statement In regard to the context.

Certainly, prob-
ably and possibly ', among others, show how much reliability
Is to be ascribed to the statement.

Perhaps It Is worth saying,
though the matter should be sufficiently obvious, that no importance should be attached to the grouping of verbs and
adverbs Into three sets which has been adopted.

It has been
done purely for convenience in an outline exposition. There
are differences between the members of each of my groups
and the groups are not sharply divided ; it is easy to think
of verbs which might with equal reason be placed In either of
two groups.

Once again it must be said that our aim is to
lay down general lines for the interpretation of parenthetical
verbs, not to do full justice to any of them.

Comparison of these adverbs and parenthetical verbs.

Provided that it is not construed as a list of synonyms, we can
couple these adverbs with parenthetical verbs as follows :


happily ----- I rejoice

unfortunately --- I regret

consequently -- I deduce -- I infer.

presumably --- I presume

 admittedly  --- I admit

---  certainly I know (cfr. "knowingly")

 probably ---- I believe -- cfr. "Unbelievably", "believably")

This Is not, Grice and Urmson repeat, a list of synonyms.

Apart from questions
of nuance of meaning the adverbs are more impersonal.

Admittedly suggests that what is said would be regarded by
anyone as an admission whereas

"I admit"

shows only the way
that the statement is to be regarded here, by the utterer.

Also it is not
possible to say that every adverb has a verb corresponding to
it which has more or less the same Import, or vice


But it does seem that these adverbs and parenthetical
play much the same role and have much the same grammatical relation to the statements which they accompany, and that,
therefore, the comparison is illuminating in both directions.

But now Grice and Urmson meet an objection which will certainly be
made by some philosophers to this comparison ; and Grice and Urmson intend
to meet it by a fairly detailed examination of the example
which they themselves would most likely choose.

In doing
this Grice and Urmson further explain the use of parenthetical verbs.

Probably and I believe.

To say, that something is probable, my imaginary objector will say, is to imply that it is
reasonable to believe, that the evidence justifies a guarded
claim for the truth of the statement ; but to say that someone
believes something does not imply that it is reasonable for
him to believe it, nor that the evidence justifies the guarded
claim, to truth which he makes.

Therefore, the objector will
continue, the difference between the use of the word ' believe'
and the word 'probably* is not, as we have suggested, merely
one of nuance and degree of impersonality, for in one case
reasonableness is implied and in the other not. This objection
can be met, but to do so we must first make a more general
point.



Implied claims to reasonableness.

There was an implied claim to truth whenever a sentence is
uttered in a standard context, and the meaning of this is
explained.

Now we must add that whenever we make a
statement in a standard context there is an implied claim to
reasonableness, and this contention must be explained.

---- cfr. Grice on 'rational' and 'reasonable' in "Aspects of reason".


Unless we are acting or story-telling, or preface our remarks with some such phrase as

"I know I'm being silly, but ..."


or,

"I admit it is unreasonable, but ..."

it is, Grice and Urmson think, a
presupposition of communication that people will not make
statements, thereby implying their truth, unless they have
some

GROUND, however tenuous, for those statements.

To
say,

"The King is visiting Oxford tomorrow."

and then, when asked why, to answer

'Oh, for no reason at all'

would be to SIN against the
basic conventions governing the use of discourse. --- Grice goes on to provide a KANTIAN justification for that, hence his amusing talk of maxims and stuff.


Therefore, Urmson and Grice think there Is an Implied claim to

REASONABLENESS

which goes with all our statements, i.e. there is a convention that we will not make statements unless we are prepared to claim and defend their REASONABLENESS.

--- cfr. Grice on 'reasonable' and 'rational', in "Aspects of reason" (the John Locke Lectures, Oxford, 1979).

With
this prolegomenon we can return to the question of the
relation of belief and probability.

Defence of Grice's and Urmson's account of belief and probability.

We can
now say, with less risk of being misunderstood, that when a
man says,

"I believe that he is at home"

or

"He is, I believe, at home"

he both IMPLIES (or 'expresses' as Grice may prefer -- since 'implicate' is TOO STRONG)  a (guarded) claim of the truth, and also Implies a claim of the reasonableness of the statement
that he is at home.

Thus, if our objector points out that
'probably he is at home ' implies, in the view of the speaker,
the reasonableness and justifiability of the statement, we may
answer that this is equally true of * believe* In the first person
present.

In such a form as *I believe that he is at home'.


What our objector has failed to do Is to notice the vast array
of situations in which the verb 'believe* is used. We will
now single out some, but only some, of these uses.

(A) Jones says,

"Smith is, I believe, at home*.

Here Jones
makes an implied guarded claim (that is the effect of adding
'I believe') to the truth and also an implied claim to the
reasonableness, of the statement *Smith Is at home*.

This is the
case already examined.

(B) Williams, reporting Jones, says ' Smith is, Jones believes, at
home*.

This is oratio obliqua, reporting Jones' parenthetical
use of the verb.

Williams, by uttering the sentence, implies the
truth and reasonableness of the statement that Jones has made
the statement that Smith Is at home

(Jones thereby implying its
truth and reasonableness with the conventional warning signal
about the evidential situation).

(C)

Smith, who has discovered that there has been a sudden railway stoppage, sees Jones making his habitual morning dash to the station, and says,

"Jones believes that the trains are working."

This is a new, and, however important,
derivative, use of the verb 'believe*.

Note that in this context Smith could NOT say:

"The trains,, Jones believes, are working."

Jones, who has probably not considered the
matter at all, Is behaving in the way that someone who
prepared to say either The trains are running ' or c I believe
that the trains are running * would behave (no doubt he
would be prepared to say one or other of these things If he
considered the matter).

We thus t in a perfectly intelligible
way, extend our use of the verb 'to believe" to those situations
in which a person behaves as a person who has considered
the evidence and was willing to say 'I believe * would consistently behave.

In this case, but in this case
alone, there is
some point in saying that the verb is
used
DISPOSITIONALLY.

But note that it is so used with reference to another use of
'believe'.

It is also noteworthy that the verb cannot be so
used in the first person present.

To say ' I believe * In this
sense is no more possible than to say

"I am under the delusion that..."

 "I believe" is always used parenthetically, though not
always purely so.

If one does recognizes that a belief that one has held is unreasonable, one either gives it up or is driven to saying

"I can't help believing".


This is psychological
history, and carries with it no claim to truth or reasonableness.



Thus we see that

"Jones believes that p"

 does not imply the
reasonableness of "p" any more than

' It seems probable to Jones that p"  does.

On the other hand, both

Probably p.

and

I believe that p.

do imply the reasonableness of p.

Thus, so
far at least as we are concerned with the well-known objection
about reasonableness, the parallel between 'probably* and
'I believe* has stood the test without difficulty.

At the risk of digression we may pause to comment on
the history of the analysis of belief.

Of old, philosophers
tried to find a primary occurrent use of 'believe' as a psychological description ; but in recent times the impossibility of
this has been amply demonstrated, and philosophers have
resorted to the so-called dispositional analysis, assuming that
if the verb does not describe an occurrence it must describe
a tendency to occurrences.

There is, as we have seen, some
point in the traditionalist reply to this that belief Is here
analysed as being the behaviour.

If any, which would con-
sistently accompany itself. A recognition that in the analysis
of belief the non-descriptive parenthetical use is primary
seems to me to illuminate and resolve this dispute.

This is all that can here be said about belief. It far from
exhausts even all the relevant considerations, but our aim is
not to examine any one parenthetical verb exhaustively;
rather it is to shed new light on them all by presenting them
as a group. I want to say the main things which may be said
about a set of verbs which are not normally considered
together as an aid to the thorough examination of each which
I do not undertake. Individually, none of these verbs can be
exhaustively treated in their capacity as parenthetical verbs
and I must not be taken as suggesting that they can.

Further consideration of the third group.

"I guess"

has
nowadays a colloquial use in which its significance is, at the
best, very indeterminate -- or as Speranza would prefer, "otiose".


But in a stricter use it serves to
warn that what is being said is a guess.

Suppose that one is
asked '

"Do you know who called this afternoon?"

J one may
answer

"No, but I guess that it was Mrs. Jones."

Even
here one is making an implied claim that it was Mr. Jones
who called and that this is a reasonable thing to say.

If one
had said

"I guess that it was Mr. Stalin"

one would have
been making a clumsy joke and not really guessing at all.

It
seems to Urmson and Grice  to be quite impossible that anyone should think
that here ' I guess ' reports any mental events or any tendency
to behave in any special way.

It is put in to show that one
is making one's statement without any specific evidence, that
it is, in fact, a guess.

What makes it a guess is not a mental
act nor a disposition to behave in any way, but, if it is a
genuine guess, its being said without any specific evidence,
and its being potentially silly or lucky, not well-based or ill-
supported. I cannot see that there is any essential difference
between guess on the one hand and know, opine, and suspect,
for example, on the other.

The epistemological situation is
more complicated in the latter set of cases, and some of them
have special quirks in their use, know being a notorious
example, but that Is all. They are essentially the same sort
of verb.

It might be worth while to compare this view of knowledge with what J. L. Austin said in his valuable paper on * Other Minds,  P.A.S. Supp. Vol. XX (reprinted in L. L., II).

Much of Urmson's approach was suggested by this paper by Austin.

Among other, less immediately relevant,
things, Austin there distinguishes a class of performatory
verbs and compares our use of know with our use of these
verbs.

In particular, he compares it with guarantee.

But
Austin is careful not to say that know is a performatory verb.

He also points out Important differences between the two
verbs. I agree that the comparison which he makes between
know and performatory verbs is just and illuminating. Paren-
thetical and performatory verbs have much in common as
against ordinary descriptive verbs. I am not therefore dis-
agreeing with Austin, but trying to locate the verb to
in a class which it was not his purpose to consider.

Relation of the parenthetical use of parenthetical verbs to
their other uses. We have now distinguished a set of paren-
thetical verbs and have made the following main points about
their parenthetical use in the first person of the present tense :

(i) They occur in the present perfect, not the continuous
tense, though their use is different from that of the
present perfect tense of verbs which have a present
continuous tense.

(ii) Though, in a wide sense, psychological verbs, they are
not psychologically descriptive.

(iii) They function rather like a certain class of adverbs to
orient the hearer aright towards the statements with
which they are associated. The ways in which they do
this may be roughly indicated as being aids to placing
the statements aright against the emotional, social,
logical, and evidential background.

(iv) There is, as when the conjoined statements are used
alone, an implied claim for the truth and reasonableness
of these associated statements.


But parenthetical verbs are not always used parenthetically.


In the first person present, to which use we have so far confined
practically all our attention. We must now say something
about their other uses. First, we may consider the positive
analogy. In connexion with point (i) above, there Is a
positive analogy, though not a very tidy one.

The analogy
seems to hold completely in the case of some verbs ; one
cannot say f ' I was believing ', * he is believing ', c I was knowing ',
*he was knowing', *I was suspecting* or 'he was suspecting'.
In the case of some other parenthetical verbs, we find a rare
and anomalous imperfect tense. For example, we can say
that you were admitting something if you were interrupted
in the middle of a statement which you were making as an
admission ; or again, we can say that someone is deducing
the consequences of a set of premisses, while he is stating a
succession of things as deductions. But these are not genuine
exceptions. In the case of another set of these verbs an im-
perfect is not so strange. At the end of an argument which
have been put forth someone might say, for example,

"All the
while you were assuming (presupposing, accepting) that so
and so."

But this is not like the imperfect tense of ordinary
verbs which report the continuance throughout a period of
some occurrence. I was not throughout the period con-
tinuously doing an act of assumption which I carefully re-
frained from mentioning.

Rather I was arguing as a man
would reasonably argue who was prepared to say, ' I assume
that so-and-so' ; that is to say, I was arguing in a way that
required so and so as a premiss if the argument was to be
valid. I ought, therefore, to be willing to state so-and-so as
a premiss.

Thus here, too, the other use has to be understood
In the light of the parenthetical use.

We must also note that, in general, these verbs can
throughout be used in parenthesis ; we can say

"Jones was,
Smith admitted, able."

This seems to be so whenever the use
Is either definite oratio obliqua or, at any rate, a fair paraphrase.

Some verbs, such as deduce and admit, seem always to be used in this way.

But others, including, as we have
already seen^ assume, prsuppse s and are

used, not of a man who has said, ( I assume (believe, pre-
suppose) ', or words to that effect, but of a man who
as a man reasonably would who was prepared to say that.

In such a use, which is a genuine descriptive use, the parenthetical Insertion (in a grammatical sense) of the verb seems to be impossible.

Continuing with the positive analogy, it seems to follow
from the above that, except in some derivative uses, parenthetical verbs are not used as psychological descriptions in
other parts of their conjugation any more than in the first
person present. And even in these derivative uses ? they seem
to describe general behaviour rather than to be specifically
mental.

The obvious negative analogy is, first, that the adverbs
can only be used to correspond to the first person present
(see point (iii) above).

But this negative analogy is only so
in a very limited way. If the adverbs did correspond exactly
to the whole conjugation of the verb, then the conjugation
would appear to be otiose. But the adverbs can be systematic-
ally correlated with the whole conjugation of the paren-
thetical verbs with the aid of the verb to seem.

This point is
illustrated by these two groups of four sentences

"I regret
that it is too late Unfortunately it is too late ; He regretted
that it was too late it seemed to him to be unfortunately
too late ; believe that it is lost It is probably lost ; He
believed that it was lost It seemed to him to be probably
lost.

The second obvious negative analogy is that in connexion
with point (iv) what is said to be supposed, regretted, believed,
etc., by others, or by oneself in the past, is not in general
implied to be true or reasonable by the speaker (there are
exceptions to this, in each case with a special reason, know
being an obvious example).

But here, again, this is exactly
what is to be expected ; while 'He believed that it was lost',
if said by me, does not imply a modified truth claim to
reasonableness by me for the statement that it is lost, it does
allege exactly that of the man to whom I refer. The same
would be true of "He suggested 1 , 'He concluded \ and so on.
The point of these verbs remains as a kind of orientating
signal* but when not in the first person present they report
the statement-cum-signal rather than making it.

Beyond this Grice and Urmson do not see their way to a general account of
the relation of other uses of these verbs to their parenthetical
use in the first person present.

Such relation is In detail
different and more or less close in the case of different verbs.
Sometimes the parenthetical use seems to be the basic use
and to be requisite for an understanding of the others ; in
other cases the illumination is of much lower candle-power.


Grice and Urmson indicated, for example, that 1 regard the
parenthetical use as basic in the case of * believe*, and how I
regard some other standard uses as being related to it.

If
the point of * I know * is, roughly, to signal complete

trustworthiness

for a statement made in the best evidential conditions, then the point of other uses of the verb may be said
with reasonable accuracy to be the assertion that somebody
else, or oneself at another time, was in a position in which he
was entitled to say 'I know 9 . This is a different, though
equally close, connexion, from that which we found in the
case of belief.

Know is a rather special case, but many
parenthetical verbs are very similar to believe in this respect.


Thus, if to say 'I presuppose that/' is to assert/ with an
indication that it is to be fitted into the logical context as an
unproved premiss, so to say ' He presupposed that/' is either
to say that he said / in that way, or, at one remove, to say
that he put forward / as one would reasonably do if he were
making a presupposition. This last use corresponds to the
use of *he believes that/' to indicate that someone is acting
as he would reasonably act if he believed in the sense that he
was willing to say 'I believe that/'. Assume is the same in
this respect.

Deduce, conclude and guess seem to be different
again.

We would never say that someone deduced / unless
we believed that he had seen that / followed, even though it
was a possible deduction and he appeared to accept /, which
is a difference from the case of belief.

On the other hand, we can say that someone deduced p, where we are ourselves
prepared to treat p as a legitimate deduction, which is a
difference from the case of knowledge.

In the of still

other verbs which have a parenthetical use In the first
present this use does not seem to be at all central, a key to
the understanding of other uses, as In the case of the verbs
above mentioned. Examples of such verbs are hear, rejoice,
expect \ It Is among these verbs that possible exceptions to
some of our generalizations are to be sought.

It is perhaps worth mentioning a fact which will make it
especially clear that too much generalization about the
relation of other uses to the first person present is not possible.


It is possible to manufacture parenthetical uses of verbs which
are not normally parenthetical , even In the first person present,
by the addition of the infinitive to say.

Thus the verb * I am
sorry that* is normally a formula of apology or of self-
reproach ; but we can convert It Into a parenthetical verb by
the addition of to say:

"He is, I am sorry to say, unwell."

* I am glad J can be treated In the same way, and so can other
verbs.

It thus becomes abundantly clear that we must not
always try to see the parenthetical use as central.

It might
also be Interesting to note at this point that we can use the
device of the infinitive to get two parenthetical verbs into
association with one sentence.

Thus 'I regret to hear that
p* combines 'I regret' and *I hear', thereby orientating the
hearer In two different ways at the same time.

Compare * I
am sorry to conclude 1 .

These points should make it clear
that one cannot generalize too much about the relation of
parenthetical to other uses of verbs.

Before attempting to draw the threads together, we will
anticipate a possible criticism.

It may be said that the grammatical feature of being used sometimes In parenthesis, In
the grammatical sense, is not a sufficient test of a verb's
parenthetical character in my sense.

"Guarantee", it may be
said, is a performatory verb, since to say 'I guarantee' is to
guarantee, not to orientate the hearer.

Similarly to say 'I bet' Is to bet, and to say * I warrant * Is to warrant.

But we
can put these verbs into parenthesis.

My answer is that we do not put these verbs into parentheses when we are using
them in a performatory way.

Tto treat

"He'll come to a bad end, I guarantee"

as a guarantee, or to ask for the odds or to cry

"Taken!"

when someone says

"He'll forget to come, I bet"

would be, as Aristotle would say, the mark of an uneducated man.

We have here another case of the borrowing
of another sort of verb for parenthetical uses.

But it must,
of course, be acknowledged that grammatical form is likely
to be, here as elsewhere, but a fallible guide to the logical
nature of a statement.

In the end, the feature of being
capable of occurring in parentheses is only a heuristic device
for picking out a certain class of verbs, which is certainly
different from, say, performatory verbs.

A little more is said
below, about the relation of Grice's and Urmson's philosophical thesis to the
grammatical fact to which 1 draw attention.

Another objection which should be anticipated will be
made on different grounds. It may be said that I have often
given the appearance of conducting a grammatical rather
than a philosophical investigation, and that, for example,
the point about the lack of a present continuous tense could
not be made in many languages.

It is true that I have been
using the grammatical features of English as a clue to philosophical points ; but one can find similar, if different, grammatical clues in other languages.

The actual point about parenthesis seems to apply to French,

and one should try
to explain why in French one says

"Je regrette de vous
informer que votre fils est mort."

"and

"Je suis desole du fait que votre fils est mort."

It would surely be out of the question
to say

"Je regrette du fait que..."

Similarly, I am informed
by those with a better command of German than Grice and Urmson had that
their  point about the similar use of parenthetical verbs and
some adverbs is reinforced by the fact that in German the
verb would often be most naturally translated by an adverb,
e.g. 1 1 regret * would often be translated most naturally by
"Meider".

The fact that one makes use of the clues given by
one's own language does not make the thesis inapplicable to
other languages which have the same devices.

There Is, of
course^ no reason to hold dogmatically that every
has devices closely similar to the use of parenthetical verbs
in English.

We may now sum up and reiterate the point of what has
been said in this paper.

It must be admitted that there are
verbs which may be said to describe a mental process,, how-
ever mental processes have -in the end to be analysed. Ex-
amples are meditate^ ponder^ worry % imagine^ and out.
In the case of all these verbs one uses the present continuous
tense to say what is happening now.

Other verbs such as
wisk) command, implore^ or like t hate^ approve^ love are
interestingly different from the above and need discussion,
but are not discussed here.

But there is another class of
verbs, different from any of the above, whose peculiarity is that
they can be used either parenthetically in the normal grammatical sense, or else followed by that* in either case with an
indicative clause.

Further, they are so used in the present
perfect tense, though not with the same dispositional force as
are the general run of verbs.

These verbs are the ones for
which I have invented the technical name of parenthetical
verbs.

They are important because they include such philosophical war-horses as know^ believe, and deduce.

I have tried to show

(i) that when these verbs are used in the first person of
the present tense, as is very clear when they occur gram-
matically in parenthesis, the assertion proper is contained in
the indicative clause with which they are associated, which
is implied to be both true and reasonable.

----

They themselves
have not, in such a use, any descriptive sense but rather
function as signals guiding the hearer to a proper appreciation
of the statement in its context, social, logical, or evidential.


They are not part of the statement made, nor additional statements, but function with regard to a statement made rather as 'READ WITH CARE' functions in relation to a subjoined
notice, or as the foot stamping and saluting can function
in the Army to make clear that one is making an official
report.

Perhaps they can be compared to such stage-directions as

'said in a mournful (confident) tone'

with reference
to the lines of the play.

They help the understanding and
assessment of what is said rather than being a part of what
is said.

Grice and Urmson show that in the case of many
important verbs an understanding of this use of the verb is
basic for a philosophical understanding of them ; other uses
of the verbs must be explained in terms of it.

(iii) It must, however, be clearly understood that there is
a great deal which importantly needs saying about these verbs
which has not been said here.

Grice and Urmson have not attempted to say
all that there is to say about know^ for instance, but only
to bring out certain peculiarities which it has in common
with a number of other verbs alongside of which it is not
normally considered.

But Grice and Urmson must not be too modest.

Grice and Urmson
have exposed such views as that these verbs report occurrences
or tendencies to behave in certain ways.

Most philosophers
have been obsessed with the idea that verbs always describe
some goings on if not a simple event, then a complicated
set of events.

Grice and Urmson have tried to pick out one class of verbs
which do not report any goings on or even patterns of goings
on at all. That the present discussion of them has been lucid
and accurate, let alone final, may very well be doubted ; that
the set of characteristics which these parenthetical verbs
share is significant and important is, however, something of
which Grice and Urmson feel very much more confident.