A Logical Model of Nash Bargaining Solution

Dongmo Zhang
Intelligent Systems Laboratory
School of Computing and Information Technology
University of Western Sydney, Australia
[email protected]
Abstract
This paper presents a logical extension of Nashs
Cooperative Bargaining Theory. We introduce
a concept of entrenchment measurement, which
maps propositions to real numbers, as a vehicle
to represent agents belief states and attitudes to-
wards bargaining situations. We show that Nashs
bargaining solution can be restated in terms of bar-
gainers belief states. Negotiable items, bargaining
outcomes and conicting arguments can then be ex-
plicitly expressed in propositional logic meanwhile
Nashs numerical solution to bargaining problem is
still applicable.
1 Introduction
Negotiation or bargaining is a process to settle disputes and
reach mutually benecial agreements. Typical situations of
negotiation are characterized by two or more agents who have
common interests in cooperating but have conicting inter-
ests in the way of doing so. The outcomes of negotiation
depend on agents attitudes towards their bargaining items
and their expectations from the negotiation. The represen-
tation of bargainers attitudes in game theory is implicit via
a pair of von Neumann-Morgenstern utility functions [Nash
1950]. The conicting of interests between bargainers is
then implied by certain mathematical conditions of the util-
ity functions. However, agents attitudes can be more explic-
itly described in terms of logic. Negotiable items, conicting
claims, arguments of disputation and negotiation protocols
are all expressible by logical statements [Sycara 1990][Sierra
et al. 1997][Kraus et al. 1998][Wooldridge and Parsons
2000][Meyer et al. 2004a][Zhang et al. 2004]. One dif-
culty of logical frameworks of negotiation is that quantita-
tive criteria are harder to be applied to bargaining processes,
which seem necessary in the analysis of negotiation situa-
tions. This paper attempts to bridge the gap between the
quantitative analysis of game-theory and qualitative descrip-
tion and reasoning of logic.
One of our basic assumption to negotiation is that out-
comes of negotiation are determined by bargainers belief
states. Representation of agents belief states is normally by
a set of beliefs and a revision operator over the belief set in
terms of the theory of belief revision [G ardenfors 1988]. Al-
ternatively, belief states can also be specied by an order-
ing over propositions, referred to as epistemic entrenchment
[G ardenfors 1988]. It is well-known that such two modelings
of belief states is equivalent. The main nding of the paper is
that the ordering of epistemic entrenchment can be extended
into a numerical measurement over propositions so that a von
Neumann-Morgenstern utility function is induced. With this
quantitative representation of bargainers belief states, Nashs
numerical requirements on bargaining solutions can be re-
stated in terms of logical properties. Bargaining items, ne-
gotiation outcomes and conicting arguments can be also ex-
plicitly expressed via the extended concept of belief states.
The structure of the paper is the following. After a short
motivation in this section, we extend the AGMs epistemic
entrenchment to a map from propositions to real numbers.
Bargaining problem is then dened as a pair of bargainers
belief states and outcomes of bargaining is specied by a pair
of concessions made by two bargainers. In order to induce a
von Neumann-Morgenstern utility function over negotiation
outcomes, the standard game-theoretical process of random-
ization over possible outcomes is applied. When all these
done, Nashs bargaining solution is ready to be restated in
terms of bargainers belief states. Finally we argue that our
solution to bargaining problem can not be replaced by the
belief revision based solution even though they share many
common logical properties.
We will work on a propositional languages L with nite
many propositional variables. A set K of sentences in L is
said to be logically closed or to be a belief set if K = Cn(K),
where Cn(K) = L : K . If F
1
, F
2
are two sets
of sentences, F
1
+ F
2
denotes Cn(F
1
F
2
). We shall use
and to denote the propositional constants true and false,
respectively.
2 Entrenchment Measurement
In the theory of belief revision, the belief state of an agent
consists of a belief set - the beliefs held by the agent - and
a revision operator - which takes a piece of new information
as input and outputs a new belief set. We know that the be-
lief state is uniquely determined by the agents epistemic en-
trenchment - an ordering on beliefs [G ardenfors 1988]. How-
ever, we shall argue that such an ordering is not enough to
specify agents belief states in bargaining. We will extend the
concept of epistemic entrenchment into a more general con-
cept so that a quantitative measurement can be used to capture
bargainers attitudes towards their bargaining items.
Denition 1 An entrenchment measure is a pair of (, e)
where e is a real number and is a function from L to the
real number set ' which satises the following condition:
(LC) If
1
, ,
n
, min(
1
), , (
n
) ().
Let K = L : () e, called the derived belief set of
the entrenchment measure. It is easy to see that K is logically
closed, i.e., K = Cn(K), so it is a belief set. Obviously, K
is consistent iff () < e. We shall call the belief state of
an agent and e is the bottom line of the belief state.
The following proposition shows that an entrenchment
measure uniquely determines an epistemic entrenchment,
thus also determines a unique belief revision function over
the derived belief set.
Proposition 1 Let (, e) be an entrenchment measure and K
be its derived belief set. Dene an ordering over L as
follows: for any , L,
1. if , iff ;
1
2. if , and K, iff () ();
3. if , K, for any .
Then satises all postulates (EE1)-(EE5) for epistemic en-
trenchment (See [G ardenfors 1988]).
Based on the result, we can dene a belief revision operator
that satises all AGM postulates: K if and only if
K+ and either () < () or .
We shall call the revision function derived from (, e).
Since our language is nite, it is easy to dene a multiple
revision operator by the singleton revision function:
K F = K (F)
where (F) is the conjunction of all the sentences in F. This
will facilitate the comparison of our results with [Meyer et al.
2004b] and [Zhang et al. 2004].
We remark that any epistemic entrenchment ordering can
be extended to an entrenchment measure while keeping the
ordering on sentences. Clearly such an entrenchment mea-
sure is not unique. An entrenchment measure contains richer
structure than an epistemic entrenchment ordering. Interest-
ingly, the extra structure of entrenchment measures cannot be
captured by belief revision operations.
Theorem 1 (Independence of Monotone Transformations)
Let be a strictly increasing monotonic transformation over
', i.e., for any x, y ', x < y if and only if (x) < (y).
Let and

be the derived revision functions from (, e) and

( , (e)), respectively
2
. Then for any , K = K

,
where K is the belief set of (, e).
3 Bargaining Problem
We shall restrict us to the bargaining situations where only
two parties are involved. A bargaining situation is a situ-
ation in which both bargainers bring their negotiable items
to the negotiation table expecting to reach a common agree-
ment. Whenever a conict presents, concessions form one or
both agents are required in order to reach a compromise. For-
mally, a bargaining situation or a bargaining game is a pair,
((
1
, e
1
), (
2
, e
2
)), of entrenchment measures over L such
that each belief set derived by the entrenchment measures is
logically consistent, i.e. K
i
= L :
i
() e
i
, .
1
Thanks to the anonymous reviewers for pointing out an error in
the original version.
2
is the composition operator on real number functions.
Example 1 Consider the following negotiation scenario: a
buyer (agent 1) negotiates with a seller (agent 2) about the
price of a product. The buyers reserve price is less than $16
and the sellers reserve price is no less than $10. Lets dis-
cretize the problem as follows:
p
1
= the price is less than $10 .
p
2
= the price is less than $12 .
p
3
= the price is less than $14 .
p
4
= the price is less than $16 .
Assume that the entrenchment measure of each agent is re-
spectively the following
3
:

1
() =
1
(p
1
p
2
) =
1
(p
2
p
3
) =
1
(p
3

p
4
) =
1
(p
4
) = 4,
1
(p
3
) = 3,
1
(p
2
) = 2,
1
(p
1
) = 1,

1
(p
j
) = 0 for j = 1, , 4.

2
() =
2
(p
1
p
2
) =
2
(p
2
p
3
) =
2
(p
3
p
4
) =

2
(p
1
) = 4,
2
(p
2
) = 3,
2
(p
3
) = 2,
2
(p
4
) = 1,

2
(p
j
) = 0 for j = 1, , 4.
Suppose that both sides set their bottom lines to be
1. Then K
1
= Cn(p
1
, p
2
, p
3
, p
4
) and K
2
=
Cn(p
1
, p
2
, p
3
, p
4
).
Note that the values of entrenchment measures reect the
degree of entrenchment on negotiable items of each agent
rather than the agents preference. For instance, the buyer
in above example entrenches p
4
more rmly than p
1
even
though p
1
is more protable because p
4
is much easier to
keep than p
1
4
.
Denition 2 A bargaining game B = ((
1
, e
1
), (
2
, e
2
)) is
said to be a subgame of another bargaining game B

=
((

1
, e

1
), (

2
, e

2
)), denoted by B _ B

, if, for each i = 1, 2,

1.
i
() =

i
() for any K
i
;
2. e
i
e

i
.
Note that B _ B

implies K
i
K

i
(i = 1, 2). Therefore B
can be viewed as a kind of conservative concession to each
other in the sense that each player gives up part of their initial
demands by raising their negotiable bottom lines while keep-
ing their entrenchment measurements on negotiable items un-
changed.
4 Possible Outcomes of Bargaining
We nowconsider the possible outcomes of a bargaining game.
Apparently the outcome of a bargaining process is the agree-
ments that are reached in the negotiation. In most situations
when the demands of two agents conict, concessions from
players are required. The nial negotiated outcome then con-
sists of the combination of those demands that each agent
chooses to retain. In this section we shall rst consider all
possible compromises a bargaining process could reach and
then combine these outcomes in a probabilistic fashion.
3
To make the presentation simple, we do not give the entrench-
ment measurements for the whole language. The readers are invited
to complete the measurements with reasonable ranking for the rest
of sentences.
4
This exactly follows the original explanation of epistemic en-
trenchment. For instance, is no less entrenched than be-
cause is easier to retain than , so keeping is less costly
than from the information economics point of view.
4.1 Pure deals of bargaining
Following [Meyer et al. 2004a] and [Zhang et al. 2004],
we shall dene an outcome of bargaining as a pair of subsets
of two agents bargaining item sets, interpreted as the con-
cessions made by both agents. Considering the real-life bar-
gaining, bargainers normally intend to keep their highly en-
trenched negotiable items and to give up those less entrenched
items if necessary. This idea leads to the following denition
of possible bargaining outcomes.
Denition 3 Let B = ((
1
, e
1
), (
2
, e
2
)) be a bargaining
game. K
i
(i = 1, 2) is the belief set derived by (
i
, e
i
). A
pure deal of B is a pair (D
1
, D
2
) satisfying:
1. D
i
K
i
2. D
i
3. D
1
D
2
,
4. if K
i
and D
i
(
i
()
i
()), then D
i
where i = 1, 2. The set of all pure deals of B is denoted by
(B).
There is little need to comment on the rst three conditions.
They are just the statements on the type of compromises de-
sired. The last one expresses the idea that the procedure
of concession by both agents is in the order from lower en-
trenched items to higher entrenched items: higher entrenched
items should be retained before any lower entrenched items
are considered to be given up. Combining this condition with
the second implies the individual rationality if an agent en-
trench an item as rmly as a tautology, this item will never be
given up in any bargaining situation. Such an item is called
reserved item.
We are now ready to dene the utility of pure deals for
each player. We will evaluate an agents gain from a deal by
measuring the length of the remained negotiable items of the
agent.
Denition 4 For any D (B), dene the utility of the
deal, denoted by u(D) = (u
1
(D), u
2
(D)), as follows:
u
i
(D) =
i
() min
i
() : D
i

Example 2 Consider Example 1 again. All possible pure

deals of the bargaining game are:
D
1
= (Cn(p
2
, p
3
, p
4
), Cn(p
1
))
D
2
= (Cn(p
3
, p
4
), Cn(p
1
, p
2
))
D
3
= (Cn(p
4
), Cn(p
1
, p
2
, p
3
))
The utility values of the deals are:
u(D
1
) = (2, 0), u(D
2
) = (1, 1), u(D
3
) = (1, 0).
Note that (Cn(p
1
, p
2
, p
3
, p
4
), Cn()) and
(Cn(), Cn(p
1
, p
2
, p
3
, p
4
)) are not pure deals
because p
4
and p
1
are reserved items of agent 1 and 2,
respectively.
Given a bargaining game B, if (B) is nonempty, it must
contain two extreme cases of pure deals, called all-or-nothing
deals:

D = (K
1
, R
2
) and

D = (R
1
, K
2
)
where R
i
= K
i
:
i
() =
i
(), i.e., the reserved
items of agent i.
The utilities of the all-or-nothing deals give the up-bound
and low-bound of utility values for each agent:
u(

D) = (
1
() e
1
, 0) and u(

D) = (0,
2
() e
2
).
We say that a pure deal D dominates another pure deal
D

, denoted by D ~ D

, if either D
1
D

1
D
2
D

2
or
D
1
D

1
D
2
D

2
. D weekly dominates D

, denoted by
D _ D

, if D
1
D

1
D
2
D

2
.
A pure deal D of a bargaining game B is said to be Pareto
optimal over (B) if there does not exist another pure deal in
(B) which dominates D.
The following theorem shows that if a negotiation function
takes a Pareto optimal pure deal as its outcome, then it satis-
es all logical requirements put on negotiation functions by
[Meyer et al. 2004a] and all the postulates except Iteration
proposed in [Zhang et al. 2004].
Theorem 2 For any (D
1
, D
2
) (B), if it is Pareto opti-
mal, then it satises all the postulates introduced by [Meyer
et al. 2004a] for so-called concession-permissible deals
5
:
(C1) D
i
= Cn(D
i
) for i = 1, 2.
(C2) D
i
K
i
for i = 1, 2.
(C3) If K
1
K
2
is consistent, D
i
= K
i
for i = 1, 2.
(C4) D
1
D
2
is consistent.
(C5) If either K
1
D
2
or K
2
D
1
is consistent, then K
1

K
2
Cn(D
1
D
2
).
Moreover, if
1
and
2
are the belief revision operators
derived by the two entrenchment measures of B, respec-
tively, then D satises the following conditions introduced by
[Zhang et al. 2004]:
1. Cn(D
1
D
2
) = (K
1

1
D
2
) (K
2

2
D
1
).
2. K
1
(K
2

2
D
1
) K
1

1
D
2
and K
2
(K
1

1
D
2
)
K
2

2
D
1
.
The following lemma shows that the utility value of a pure
deal uniquely determines the deal. The fact will be used in
the next section.
Lemma 1 For D, D

(B),
u(D) = u(D

) iff D = D

.
4.2 Mixed deals
In many cases, a bargaining game could end in a tie that no
single agreement reaches. A standard method to deal with
the problem in game theory is to play a lottery to determine
the outcome of a bargaining game. Following Nashs utility
theory, we call rD

+ (1 r)D

a mixed deal if D

and D

are two pure deals or mixed deals, meaning the lottery which
has the two possible outcomes, D

and D

, with probabilities
r and 1r, respectively. The set of all mixed deals is denoted
by (B)
6
.
For any mixed deal D = rD

+ (1 r)D

, the utility of
the deal can be dened as
u(D) = ru(D

) + (1 r)u(D

).
The concept of domination then can be extended to mixed
deals: for any mixed deals D, D

(B),
D ~ (_) D

iff u(D) > () u(D

).
7
5
In [Meyer et al. 2004a], six postulates were introduced to
specify the rationality of concessionary. However the sixth, (C6),
holds trivially provided K1 and K2 are logically closed.
6
More precisely, (B) is a conservative extension of (B) by
allowing lottery deals in the form rD

+ (1 r)D

.
7
For any x, y
2
, x > y denotes x1 > y1 x2 y2 or
x1 y1 x2 > y2. x y denotes that xi yi.
Since the language we consider is nite, the set of pure
deals of a bargaining game is also nite. The following
lemma is easy to prove according to the construction of mixed
deals.
Lemma 2 Let D
1
, , D
m
(B) be all pure deals of a
bargaining game B. Any mixed deal D (B) can be ex-
pressed as a linear combination of D
1
, , D
m
, i.e., there
exist real numbers,
1
, ,
m
, satisfying
1.
j
0, j = 1, , m;
2.

j
= 1;
3. D =

j
D
j
.
Moreover, u(D) =

j
u(D
j
).
In other words, (B) is a convex hull of (B) if we iden-
tify a deal with its utility pair. Obviously, the utility function
over mixed deals dened above is a pair of von Neumann-
Morgenstern utility functions.
We have seen from Lemma 1 that a pure deal can be
uniquely determined by its utility values. However, a deal can
be duplicately represented in the form of mixed deals. For in-
stance, any pure deal D can be represented as rD+(1r)D.
To solve the problem we can either use Lemma 2 to obtain
a unique representation of a deal by orthogonalizing repre-
sentation of pure deals (or utility pairs) or apply equivalence
classication by dening an equivalent relation: D D

iff
u(D) = u(D

). To avoid too much complexity, we omit the

details of the mathematical treatment but simply assume that
for any D, D

(B), D D

iff u(D) = u(D

).
5 Bargaining solution
The target of bargaining theory is to nd theoretical predic-
tions of what agreement, if any, will be reached by the bar-
gainers. John Nash in his path-breaking papers introduced
an axiomatic method which permitted a unique feasible out-
come to be selected as the solution of a given bargaining
problem[Nash 1950][Nash 1953]. He formulated a list of
properties, or axioms, that he thought solution should satisfy,
and established the existence of a unique solution satisfying
all the axioms. In this section we shall give a brief summary
of Nashs theory and then extend Nashs result in terms of
bargainers belief states.
5.1 The Nash bargaining solution
A bargaining game in Nashs terminology is a pair (, d),
where '
2
is a set of possible utility pairs and d '
2
the disagreement point. It was assumed that is convex and
compact. All bargaining games satisfying the conditions are
collected in the set B. A bargaining solution is then a func-
tion f : B '
2
such that f(, d) . Nash proposed
that a bargaining solution should possess the following four
properties [Nash 1953]
8
:
PO (Pareto-Optimality) There does not exist (u
1
, u
2
)
such that (u
1
, u
2
) > f(, d).
SYM (Symmetry) If (, d) is symmetric, then f
1
(, d) =
f
2
(, d).
8
See more modern treatments and detailed explanation from
[Roth 1979][Owen 1995][Houba and Bolt 2002][Napel 2002].
IEUR (Independence of Equivalent Utility Representation )
Let (

, d

) be obtained from (, d) by a strictly increas-

ing afne transformation (x) = (
1
(x
1
),
2
(x
2
))
9
.
Then f(

, d

) = (f(, d))
IIA (Independence of Irrelevant Alternatives) If

and
f(, d)

, then f(

, d) = f(, d).
Theorem 3 (Nashs Theorem) A bargaining solution f sat-
ises all the four properties if and only if f = F, where F is
dened by F(, d) = u

where u

solves the maximization

problem maxu
1
u
2
: u = (u
1
, u
2
) , u d.
5.2 Bargaining solution with belief states
We might have noticed that a solution to bargaining problem
is to provide a general model of the bargaining process rather
than a selection function that picks up a point as outcome
from all feasible solutions for a particular bargaining situa-
tion. Asurprise with Nashs theoremis that the axioms shown
above seemso natural that any negotiation process should fol-
low whereas the solution is so specic that it uniquely deter-
mines an outcome in any bargaining situation. The question
now is that whether we can restate these plausible require-
ments in terms of bargainers belief states instead of utility
functions.
Given a nite propositional language, let B denote all
bargaining games in the same propositional language with
nonempty set of deals. A bargaining solution is dened as a
function f which maps a bargaining game B B to a mixed
deal in (B). Following Nashs approach, we propose the
following axioms:
PO (Pareto Optimality) There does not exist a deal D
(B) such that D ~ f(B).
MD (Midpoint Domination) f(B) _
1
2

D +
1
2

D.
IEEM (Independence of Equivalent Entrenchment Measure-
ment) Given a bargaining game B = ((
1
, e
1
), (
2
, e
2
)),
for any strictly increasing afne transformations ,
f((B)) f(B)
where (B) = ((
1

1
,
1
(e
1
)), (
2

2
,
2
(e
2
))).
IIA (Independence of Irrelevant Alternatives) If B

_ B and
f(B) (B

), then f(B

) f(B).
We can easily see the similarities between the above ax-
ioms and Nashs. The differences are also observable.
Firstly, the above axioms are presented in terms of bargain-
ers belief states, i.e., the entrenchment measurement of two
agents. This is signicant not only because bargainers be-
lief states contains the logical implication of negotiation de-
mands but also because the entrenchment measurements are
more obtainable than the utility functions over all possible
outcomes(combinatorial explosion can happen with possible
outcomes). Secondly, our axioms are more intuitive and eas-
ier to verify. For instance, the MD, which is a replacement of
Nashs SYM, expresses the idea that a minimal amount of co-
operation among the players should allow them to do at least
as well as the average of their preferred outcomes. When no
easy agreement on a deterministic outcome is obtained, the
tossing of a coin is always an option to determine a winner
9
A strictly increasing afne transformation is a pair of linear real
functions with positive slope, i.e., i(x) = aix +bi where ai > 0.
of the negotiation (see a counterpart of MD in game theory in
[Thomson 1994]). Additionally, our IIA is much weaker than
Nashs IIA because not every convex subset of '
2
can be a
set of utility pairs of mixed deals.
Exactly like the Nashs theorem, there exists a unique bar-
gaining solution possessing the above properties.
Theorem 4 There is a unique solution f which satises PO,
IEEM, MD and IIA. Moreover, the solution is the function
F(B) D

such that D

(B) and u
1
(D

)u
2
(D

) =
max
D(B)
(u
1
(D)u
2
(D)).
Proof: The whole proof of the theorem is quite lengthy. To save
space, we shall only present a sketch of the second half of the
proof. We shall prove that if f is a bargaining solution possessing
PO,IEEM, MD and IIA, then f = F.
Given a bargaining game B = ((1, e1), (2, e2)), let Ki be the
derived belief sets of B. In the case that K1 K2 is consistent, it is
easy to show that f(B) F(B) (K1, K2). So we assume that
K1 K2 is inconsistent.
Let D

F(B). Suppose that B

is the bargaining game derived

from B by changing its entrenchment measurement with a strictly
increasing afne transformation so that u

(D

) = (
1
2
,
1
2
). By
IEEM, f(B) D

iff f(B

) D

. Since F satises IEEM, D

also maximizes u

1
(D)u

2
(D) over (B

). Therefore we are left to

prove f(B

) D

.
Let B

= ((

1
, e

1
), (

2
, e

2
)). Following Nashs argument, it
is not hard to verify that for any D (B

), (u

1
(D), u

2
(D))
{(x, y) : x 0, y 0 and x + y 1} because (B

) is convex.
Particularly,

i
() e

i
1.
Since K1K2 is inconsistent and (B) is nonempty, we can pick
up two Pareto optimal pure deals from (B

) in the form (K1, U2)

and (U1, K2). Construct a game B

= ((

1
, e

1
), (

2
, e

2
)) as fol-
lowing: for each i = 1, 2,
1. for any Ki,

i
() =

i
();
2. for any Cn(Ki {(Ui)})\Ki,

i
() =

i
() 1;
3. e

i
=

i
() 1.
It is easy to prove that B

is a subgame of B

and
K

i
= Ki {(Ui)} (i = 1, 2). According to MD,
u

(f(B

)) u

(
1
2

+
1
2

D

) = (
1
2
,
1
2
). On the other hand,
for any mixed deal D (B

), by Lemma 2, D

jD
j
,
where D
j
(B

). For each j, if both D

j
i
Ki (i = 1, 2), then
u

(D
j
) = u

(D
j
); otherwise either D
j
=

D

or D
j
=

D

,
which implies u

1
(D
j
) +u

2
(D
j
) = 1. Thus u

1
(D) +u

2
(D) 1.
This implies u

1
(f(B

)) + u

2
(f(B

)) 1. Therefore
u

(f(B

)) = (
1
2
,
1
2
). It follows that f(B

) is expressible in B

, or
f(B

) (B

). According to IIA, f(B

) f(B

). We conclude
that f(B) f(B

) D

F(B).
With the theorem we can easily see that the so-
lution to the bargaining situation described in Exam-
ple 1 is the equivalent class of the pure deal D
2
=
(Cn(p
3
, p
4
), Cn(p
1
, p
2
)), under which both agents
agree that the price of the product is less than $14 but no less
than $12. Note that a more rened discretization of the prob-
lem might give a mixed deal as the solution of the bargaining
game.
The uniqueness of bargaining solution depends on the as-
sumption that both bargainers agree to randomize between
outcomes, i.e., whenever the solution of a bargaining game
gives a mixed deal, a lottery will be played to decide which
pure deal will be the agreement. However, the assumption
is not generally applicable to most of real-life bargaining. In
this case, we could use a pure deal that maximizes the prod-
uct of utility pairs as an approximate prediction of bargaining
outcomes even though it is not necessarily unique (Note that
such a solution must be Pareto optimal).
6 Discussion
We have observed that a bargaining game consists of two in-
dependent entrenchment measurements on negotiable items
of two individuals. The actual values of entrenchment mea-
sures can be freely scaled without changing the solution
of the bargaining game. How could these random assign-
ments of an entrenchment measure determine a unique solu-
tion? Before we answer the question, lets explore another
example.
Example 3 [Owen 1995] Two men are offered $100 if they
can decide how to divide the money. The rst man is assumed
to be very rich, while the second man has $50 capital in all.
As the rst player is very rich, it is assumed that the utility of
$x for the rst player, with 0 x 100, is proportional to
x, i.e. u
1
(x) = x. To the second player, it is assumed that
the utility of a sum of money is proportional to its (natural)
logarithm, i.e., u
2
(x) = ln(150 x) ln 50.
The game-theoretical solution to the problem is sim-
ply maximizing the product of two agents utilities :
u
1
(x)u
2
(x) = xln(
150x
50
), which gives approximately x =
57.3, meaning that the rst player receives $57.3 and the other
takes $43.7.
The logical solution to the problem needs a process of dis-
cretization. Let P(x) denote the proposition that the rst
player receives no less than $x, where x is a natural number
(0 x 100)
10
.
The negotiable item set of each agent is then:
K
1
= Cn(P(x) : 0 x 100) and K
2
=
Cn(P(x) : 0 x 100).
Let C = P(x) P(x 1) : 0 < x 100, which
is the common knowledge of both players. The associated
entrenchment measure of the rst player can be dened as:

1
() = 100 for each C;

1
(P(x)) = 100 x for any 0 x 100;

1
() = 1 for any , K
1
.
The entrenchment measure of the second player is:

2
() = 100 for each C;

2
(P(x)) = 100 ln(
150x
50
) for any 0 x 100;

1
() = 1 for any , K
2
.
It is easy to verify that the bargaining solution of the game
is (Cn(P(57)C), Cn(P(58)C)), which gives the
similar result as the game-theoretical solution.
The result seems strange: the poor man receives less
money than the rich man. This is because the poor man is so
eager to get money that he highly entrench each single dollars
(98
2
(P(x)) 100 for any 0 x 100))
11
. If the
second agent linearly entrenched its gain as the rst agent, the
10
Note that we treat P(x) as a proposition rather than a predicate.
11
Game-theory explains such a phenomenon as the interpersonal
differences of attitudes towards threats or negotiation power of dif-
negotiation would end with 50 to 50. This means that the non-
linearity of the second agents entrenchment measurement re-
sults the non-balanced allocation. In other words, the distri-
bution of entrenchment measurement reects the bargaining
power of players. Note that the logarithm function is strictly
monotone. Therefore the non-linearity of entrenchment mea-
surement cannot be captured by any belief change operations
(see Theorem 1). The AGMs entrenchment ordering is not
enough to measure players bargaining power. In fact, the en-
trenchment measurement plays two rules in bargaining: its
ordering determines a unique belief revision operation for
minimizing the loss of bargainers negotiable items and the
distribution of entrenchment values decides the negotiating
power of bargainers.
7 Conclusion and Related Work
We have presented a logical framework for Nashs bargaining
solution. To do so, we introduced a concept of entrenchment
measure, which maps a proposition to a real number, and used
it as a vehicle to convey agents belief states in different bar-
gaining situations. We have shown that Nash bargaining solu-
tion can be restated in terms of the extended concept of belief
states. Negotiable items, bargaining outcomes and conicting
arguments can then be expressed in propositional logic mean-
while Nashs numerical solution to bargaining problem is still
applicable. This offers a combinative approach of qualitative
and quantitative analysis to bargaining situations. As a direct
application of the result, a logic-based solution to automated
negotiation can be proposed. Given a two-agent bargaining
situation, we invite both agents to describe their negotiable
items in terms of propositional logic and to provide their own
measurements of entrenchment on their negotiable items. All
the information comes to an arbitrator (or a server) who will
announce the outcome of the negotiation after conducting a
process of belief revision and a calculation of Nashs solu-
tion. This procedure should be considered to be fair because
the result of negotiation is in effect determined by the partic-
ipants belief states rather by the arbitrator.
This paper is closely related to the work on the belief
revision model of negotiation [Booth 2001][Meyer et al.
2004a][Meyer et al. 2004b][Zhang et al. 2004]. We have
proved that if the bargaining solution takes a pure deal as the
outcome, it will satisfy almost all assumptions that are put
on the negotiation function constructed by belief revision op-
erators(see Theorem 2). However, as we have discussed in
the last section, belief revision operation cannot measure bar-
gainers negotiation power. Therefore a negotiation function
dened purely by belief revision operation is unable to deter-
mine negotiation outcomes.
The model we presented in the paper is a logical exten-
sion of Nashs cooperative model of bargaining, which deals
with the bargaining situations described by a set of abstract
rules. Non-cooperative models of bargaining, in contrast, an-
alyze interaction which is based on explicit rules of bargain-
ing [Binmore et al. 1992]. A series of work have been done
that attempts to represent negotiation procedures in terms of
ferent negotiators [Owen 1995][Houba and Bolt 2002][Fatima et al.
2005].
logic [Kraus et al. 1998][Sycara 1990][Sierra et al. 1997].
It is interesting to see whether the classical results about the
relationship between cooperative solutions and noncoopera-
tive solutions of bargaining problem can be extended to the
logical models of bargaining.
Nashs axiomatic approach to bargaining has reached a
high sophistication through the development of last ve
decades [Roth 1979][Thomson 1994][Napel 2002]. A vari-
ety of alternative assumptions have been proposed to derive a
given solution concept. It could be a promising research topic
to investigate logical properties of these alternative solutions.
On the other hand, the logical implication of the distribution
of entrenchment measurements is also worthwhile to be fur-
ther explored.
References
[Binmore et al. 1992] K. Binmore, M. Osborne and A. Ru-
binstein, Noncooperative models of bargaining, Hand-
book of Game Theory, Vol 1, 180-225.
[Booth 2001] R. Booth, A negotiation-style framework for
non-prioritized revision, TARK01, 137-150.
[Fatima et al. 2005] S. Fatima, M. Wooldridge and N. R.
Jennings, N. R., Bargaining with incomplete information,
to appear in Annals of Mathematics and Articial Intelli-
gence.
[G ardenfors 1988] P. G ardenfors, Knowlege in Flex: Mod-
eling the Dynamics of Epistemic States, The MIT Press,
1988.
[Houba and Bolt 2002] H. Houba and W. Bolt, Credible
Threats in Negotiations, Springer.
[Kraus et al. 1998] S.Kraus, K.Sycara, A.Evenchik, Reach-
ing agreements through argumentation: a logical model
and implementation, Articial Intelligence, 104, 1-69.
[Meyer et al. 2004a] T. Meyer, N. Foo, R. Kwok and D.
Zhang, Logical foundations of negotiation: outcome,
concession and adaptation, AAAI-04, 293-298.
[Meyer et al. 2004b] T. Meyer, N. Foo, R. Kwok and D.
Zhang, Logical foundations of negotiation: strategies and
preferences, KR04, 311-318.
[Nash 1950] J. Nash, The bargaining problem, Economet-
rica, 18(2), 155-162.
[Nash 1953] J. Nash, Two-person cooperative games,
Econometrica, 21(1), 129-140.
[Napel 2002] S. Napel, Bilateral Bargaining, Springer.
[Owen 1995] G. Owen, Game Theory, Academic Press.
[Roth 1979] A. Roth, Axiomatic Models of Bargaining,
Springer-Verlag.
[Sierra et al. 1997] C. Sierra, N. R. Jennings, P. Noriega, and
S. Parsons, 1997. A framework for argumentation-based
negotiation,ATAL-97, 177-192.
[Sycara 1990] K. Sycara, Persuasive arguementation in ne-
gotiation, Theory and Decision, 28:203-242.
[Thomson 1994] W. Thomson, Cooperative models of bar-
gaining, Handbook of Game Theory, Volume 2, Elsevier
Science B.V., 1237-1284.
[Wooldridge and Parsons 2000] M. Wooldridge and S. Par-
sons, Languages for negotiation, ECAI2000, 393-400.
[Zhang et al. 2004] D. Zhang, N. Foo, T. Meyer and R.
Kwok, Negotiation as mutual belief revision, AAAI-04,
317-322.