The future of paraconsistent logic

Jean-Yves Beziau
Contents
1. Logic of the future or logic without future ?
2. Paraconsistent problems
2.1. The problem of denition of paraconsistent logic
2.2. Philosophical problems
2.3. Mathematical problems
3. Relations between paraconsistent logic and other logics
3.1. Fuzzy logic and many-valued logic
3.2. Relevant logic
3.3. Modal and intensional logics
3.4. Intuitionistic logic, paracomplete logics and paranormal logics
3.5. Non monotonic logic, alfabar logic, substructural logics
3.6. Non Frego-Aristotelean logic
4. Applications of paraconsistent logic
4.1. Applications to human sciences
4.2. Applications to natural sciences
4.3. Applications to formal sciences
4.4. Applications to cyber sciences
5. Paraconsistent logic and the third millenium
6. References
1 Logic of the future or logic without future ?
Stefan Zweig has written a book called Brazil, country of the future, it does not
seem that his prophecy has already been fullled and there is a typical Brazilian
joke saying that Brazil is the country of the future and will stay so forever.
Maybe one can say a similar thing of paraconsistent logic. After its creation,
nearly 50 years were necessary to organize a world congress on paraconsistency
and there is still no textbook dedicated to it. For comparison, linear logic,
only ten years after its creation, is already studied worldwide, and many linear
congresses have been organized. But it is dicult to know what will happen.
Maybe paraconsistent logic will last and linear logic is just a fashionable logic
that nobody will remember in ten years. Or maybe paraconsistent logic will
1
stay a controversial curiosity and linear logic will turn into the ocial logic,
replacing classical logic, ruling our computers, wifes and money.
Or maybe paraconsistent logic and linear logic are only variations on the
same theme as classical logic. Maybe in the future there will be more radical
changes, a logical revolution similar to the Fregean one. A new conception of
logic will take place relatively to which classical logic will look like syllogistic
looks today relatively to classical logic, and paraconsistent logic and linear logic
will look like the numerous variations and alleged improvements of syllogistic
that were proposed during the 2000 years in-between Aristotle and Frege, that
nobody, apart from historians and philologists, is interesting in today.
2 Paraconsistent problems
2.1 The problem of denition of paraconsistent logic
Paraconsistent logic?
Roughly speaking, a paraconsistent logic is a logic rejecting the principle of
non contradiction (PNC). The question is whether a negation not obeying PNC
is still a negation. For some people like Quine [82], and more recently Slater
[87], the answer to this question is a big NO because negation must by denition
obey PNC. But if one thinks two minutes about the problem, he will see that
it is not that simple.
What is the denition of negation? The implicit claim of Quine and Slater
is that there is only one negation, which is perfectly and precisely described by
classical formal logic. However, on the one hand it is dicult to argue that this
classical negation gives an account of the use of natural languages negation and
therefore such denition appears to be not very descriptive but rather strongly
normative, not to say dictatorial. On the other hand, even in mathematical
logic, since many years people call negations various connectives close to, but
dierent to classical negation.
People may argue that PNC is one fundamental property of negation, and
that it makes no sense to still use the word negation to denote a connective
which does not enjoy this property. But the principle of excluded middle is also
a fundamental property of negation, a kind of dual of PNC, and intuitionistic
negation which does not enjoy this property is nevertheless called a negation.
So why one should not be allowed to call a negation an operator which may
appear as a dual (cf. Section 3.4.) of intuitionistic negation?
The question of whether a paraconsistent negation can rightly be called a
negation is still an important open problem which has two sides: 1) On the
mathematical side we must investigate which are the properties of negation
which are compatible with the rejection of PNC. 2)On the philosophical side,
we must see if these properties are enough to characterize the idea of negation.
There are still a lot of work to do on both sides. On the mathematical side the
2
question is not so simple because properties of negation are of dierent types
and levels and they depend on the framework in which we are working and
there are several possible frameworks (standard Tarskian consequence operator,
substructural sequent calculus, etc.). What is needed here to properly conduct
such investigations is a general theory of logic, a universal logic, which will play
the same role as universal algebra plays in algebra (cf [18], [21], [31]). On the
philosophical side, there also various possible ideas or theories of negation which
are not compatible and which belong to dierent contexts.
Paraconsistent logic?
As it is known, the expression paraconsistent logic was proposed in 1976
by the Peruvian philosopher Miro Quesada as a possible solution to a request
by da Costa. Very quickly the name became widespread and this name even
play a decisive role in the propagation of its denotation (cf. [43], [49]).
However there are many people who dont like it and have tried to replace
it, without success, by something else : dialethic logic, antinomic logic, parain-
consistent logic, logic of contradiction, teratologic, transconsistent logic, etc...
The Greek prex para has mainly three meanings:
- (1) against as in the expression paradox (which means in Greek, against
the common sense),
- (2) beyond as in the expression paranormal,
- (3) very similar (connected, nearby) as in the expression paramili-
tar.
This perx is therefore itslef contradictory in the sense that it has dier-
ent incompatible meanings. Some people reject the expression paraconsistent
logic because they interpret it according to the the third meaning of para and
they think that such interpretation does not match the revolutionary purpose of
paraconsistent logic which is to take us not nearby but beyond consistency. The
people who suggest the expression parainconsistent logic think that paracon-
sistent logic is beyond, or nearby inconsistency, not consistency.
What is less known is that the prex para exists in other languages with
other meanings. In Puppy-Guarana, one of the many disappeared languages of
the indians of South-America, it means out of. Although Puppy-Guarana is
not anymore spoken, many proper names in Brazil are from this language, like
the name of Rios beach Ipanema, or like the name of the city Parati which
means out of you, or like the name of the State where da Costa was born,
Parana, which means out of Na, Na being the name of a famous native
Goddess (cf. [36]).
The expression paraconsistent logic is nowadays well-established and it
will make no sense to change it. It can be interpreted in many dierent ways
which correspond to the many dierent views on a logic which permits to reason
in presence of contradictions. Nevertheless one may want to have further names
to name these views. Interpreting the prex para in its native sense, along
the line which leads from left to right, from the conformist comfortably sitted
3
on the rm dogma of consistency to the nihilist standing up, holding high the
burning ag of inconsistenty, we have the following picture:
consistent, neoconsistent, transconsistent, anticonsistent, paraconsistent, para-
inconsistent, antiinconsistent, transinconsistent, neoinconsistant, inconsistent.
2.2 Philosophical problems
Some people believe in True Contradictions and for that reason defend the
existence of paraconsistent logic. Other people believe in God and for that
reason defend the existence of the Church. In both cases the diculty is to
know what is exactly the object of their belief. And to deduce the existence
of True Contradictions from paraconsistent logic seems to be possible only by
a reasoning similar to the one with which we will deduce the existence of God
from the existence of the Church, reasoning common among people from the
Church, but rare among philosophers, Kant being the exception that proves the
rule.
In some sense we are surrounded by contradictions. But maybe these contra-
dictions are just apparent or maybe they really exist but are things that must be
avoided. In any case it seems that paraconsistent logic is essential, because even
if contradictions are only apparent, perhaps we cannot escape from the world
of appearances, we have to stay in Platos cave dealing with contradictions, and
even if we can or must avoid them, perhaps paraconsistent logic is a smoother
way than classical logic to do so.
Some people think that paraconsistent logic is essential because contradic-
tions are essential. They think that there are true contradictions that cannot
be avoided or more than this : contradiction is the essence of reality.
Clearly there is a ght over contradictions. Some people are against them,
they think that contradictions entail confusion and that they must be banished
from rational discourse, from political rhetoric and scientic method. On the
other hand there are people who think that contradictions are everywhere, in
the opposition between day and night, good an evil, truth and falsity and that
these couples are the very generating forces of reality. That life results from the
copulation between God and Satan, that there is no happiness without suering,
no truth without falsity.
People who love contradictions think that the straigth distinction between
falsity and truth is too articial, people who hate them think that confusion
between truth and falsity must be avoided because it favours falsity rather than
pure white truth. On both sides people are more interested to say what they
think contradiction is rather than to try to understand what it is, i.e. on both
sides there are few true philosophers. This is conrmed by what people write
on the subject which is a blend with a low philosophical density.
Let us take just one example. Some people say that the liar paradox proves
that a natural language such as English is contradictory and that this proves
the existence of true contradictions. But I can write It is raining over Paris
4
and it is not raining over Paris, this will not entail that 2+3=7 or that God
is Satan. And even if I say it, the earth will not explode. We must confuse
language neither with reality nor with reasoning.
People may argue that if it does not appear as a contradiction, it is because
we are taking it at a poetical level. Taking it as its right level, the assertorical
(vs. poetical) level, together with a correspondance theory of truth, it really
entails that the earth is a big apple pie, and since it does not seem so, one has
to admit that it is not possible that it is raining over Paris and it is not raining
over Paris, or one has to use paraconsistent logic. But why should we take it as
the assertorical level? Maybe the liar paradox is just a product of the poetical
power of natural language.
The future of paraconsistent logic will depend on a right analysis of the
various dierent aspects of contradiction. People saying that there are con-
tradictions everywhere, without any rational support, and that paraconsistent
logic is therefore our salvation, are similar to people alerting about a terrible
disease in order to sell a wonderful drug especially made to cure this disease.
But maybe there is no disease and the drug is a placebo. A group of people
can get power by saying that they can control dangerous creatures that are the
product of their imagination. This remembers the story of religions frightening
people with devils in order to convert them to their Churches.
By contrast to such kind of religiously oriented philosophical approaches to
paraconsistency, mathematical logic can play a fundamental role in the philo-
sophical future of paraconsistent logic, positively by constructing mathematical
systems in which there are interesting paraconsistent negations or negatively by
showing that there are no such systems.
2.3 Mathematical problems
Looking for a good paraconsistent logic
One thing is the idea of paraconsistent logic, another thing is a mathematical
system corresponding to this idea.
Obviously, Vasiliev had the idea of paraconsistent logic, but he didnt built
a system of paraconsistent logic (cf. [7], [12]). Since then many paraconsistent
systems have been proposed, but all of them have some basic defects, so that
it is not clear at all if paraconsistent logic really exists in the sense that there
exists a real paraconsistent logic. But what is a real paraconsistent logic ?
We can propose the following conditions:
- Negative criterium (rejection of the ex-falso sequitur quod libet)
- Positive criterium (strong properties which allow to speak about a negation)
- Good intuition
- Nice mathematical features.
There are several properties a negation can have, the question is to know
how one can choose a set of desired properties and build a system having those
5
properties and only those. Mathematics can put an end to some dreams, showing
that it is not possible to build such or such system.
A systematic mathematical study of negation can show which properties are
compatible with each other and which collection of properties can be unied in
a paraconsistent sysem. Such systematic study has not yet been carry out very
far. Most paraconsistentists have either a constructivist attitude, trying to build
their own better edice in the paraconsistent town, or either a deconstructivist
attitude, trying to destroy other systems nding malignant hidden defects. It is
a pity because several general results about paraconsistent negations are more
signicative that such or such ill-constructed system.
Unfortunately these general results are not very well-known, not because
they are dicult to understand but maybe because they are negative. For
example it is very easy to show (cf. [19]) that all the forms of contraposition
and reductio ad absurdum are incompatible with the idea of paraconsistency,
if we take these principles relatively to the consequence relation (e.g. if a b
and a b, then a ; if a b then b a). Weak forms of these principles
hold for negations such as intuitionitic and minimal negations. In the case of
paraconsistent logic, such principles are valid only for a kind of implications
which have very weak properties.
Paraconsistency and self-extensionality
Anyway there are some partial results related to such and such system or of
a wider width. It seems that the main problem is connected to the compatibility
between the replacement theorem and basic properties of negation. Following
Wojcickis terminology (cf. [99]), let us call self-extensional a logic in which the
replacement theorem holds. The logic C1 is not self-extensional (see [53], [54]).
Mortensen [70] has shown moreover that there are no non trivial congruences
relations denable in C1. However it is possible to develop extensions of C1 in
which there are some non trivial congruence relations (see [71], [50]. But Urbas
has shown that C1 admits no non trivial self-extensional extensions other than
classical logic [93].
We dont think that self-extensionality is necessarily a mathematical defect
for a logic (cf. [24]). But from the philosophical point of view there must be an
intuition supporting the non self-extensional behaviour of a negation.
In C1, a b is logically equivalent to b a, but (a b) and (b a) are
not logically equivalent. In LP, a a is logically equivalent to b b but
(a a) and (b b) are not logically equivalent. And in both cases there
are no intuitive justications of these abnormalities.
One has to nd a non self-extensional paraconsistent logic in which such kind
of abnormalities do not appear, or to work with self-extensional paraconsistent
logics. Let us note the intrinsic limitations of negation in a self-extensional
paraconsistent logic. It has been shown (see [26]) that such a negation cannot
obey simultaneously the double negation law and the principle of non contra-
diction in the form (a a). Consequently a self-extensional logic in which
6
all De Morgan laws hold and in which moreover holds the principle of excluded
middle cannot be paraconsistent.
Examples of self-extensional paraconsistent logics are the logic P1 [84], the
logic LDJ [95] and the logic Z [32]. P1 is an atomical paraconsistent logic in the
sense that only atomic formulas may have a paraconsistent behaviour. LDJ is a
classical paraconsistent logic (cf. [90]) in the sense that it admits all theorems of
classical logic. Atomical paraconsistent logic and classical paraconsistent logic
are both dicult to defend philosophically. But even if one can nd serious
philosophical arguments in favour of atomical paraconsistency or classical para-
consistency, the weak point of these logics is maybe that they delimit big logical
domains for paraconsistency but the behaviour of paraconsistent negations in
these regions is not strong enough.
The problem of a big paraconsistent logic
Many non classical logics are, at the propositional level, funny toys which
work quite good, but when one wants to extend them to higher levels to get a
real logic that would enable one to do mathematics or other more sophisticated
reasonings, sometimes dramatic troubles appear. This is for example the case of
modal logic. In the case of paraconsistent logic, it is dicult to have an opinion,
since the problems are not even solved yet at the propositional level.
If one extends a non self-extensional propositional logic like C1, one will
meet at the higher levels the same problems that were found at the lower level,
that can get worst or not, and new problems may also appear. In the case of
C1, in order to avoid further new problems one has to be cautious. In the usual
construction of the language of rst-order logic, two formulas like xx and
yy are dierent but they are logically equivalent. In the non self-extensional
rst-order version of C1, in order to identify these two formulas, one as to put
an ad hoc extra-axiom [42], or to modify the construction of the language using
for example Bourbakis square [22] ; in this latter case these two formulas are
logically equivalent because they are one and the same formula. To deal with non
self-extensionality one has also to modify the notion of isomorphism in order
to preserve the concept of identity (cf. [23]). Isomorphism must be dened
as preserving the structure not only at the atomic level (aRb i aRb) but
also at all levels ((a
1
, ..., a
n
) i (a
1
, ..., a
n
)). Further problems appear with
denitions [48], if for example one wants to introduce in set theory constants
such as or in the formal language. All these problems are related to non
self-extensionality and therefore appear in all non self-extensional logics, be it
C1, or J3 or LP. In some cases, where there are some congruent connectives
within the language (C1+ or Buchsbaums logics [37]), it is a little simpler but
the basic problems persist.
What happens in the case of a self-extensional paraconsistent logic? In the
case of the paraconsistent logic Z, one is confronted for its possible extension
to higher levels to the same problems that appear in modal logic, these prob-
lems maybe regarded as negligible side-eects if we dont focus on the modal
7
aspects of this logic. It seems dicult to use a classical paraconsistent logic, like
the rst-order version of LDJ, for developing paraconsistent mathematics, since
mathematics is rather about theorems and this kind of paraconsistent logic have
the same theorems as classical logic. The situation of atomical paraconsistent
logic is not better since contradictions when they pop in mathematics rather
pop at a high level of complexity. Even a very simple contradiction like Russells
paradox is not an atomical contradiction. Maybe a paraconsistent logic suitable
for mathematics should be a complexical paraconsistent logic where formulas
can have a paraconsistent behaviour only above a given degree of complexity
(see [33]).
Paraconsistent set theory and Curry-Moh Shaw Kweys paradox
The Curry-Moh Shaw Kweys paradox shows how to derive anything from
the unrestricted axiom of abstraction with the help of very few additional log-
ical principles and in particular no principles concerning explicitly negation.
Therefore paraconsistent logic is not really a possible alternative, for escaping
Russells paradox, to ZF or other classical set theories based on the restriction
of the axiom of abstraction.
However paraconsistent logic permits to deal with set theory with a restricted
axiom of abstraction together with a Russellian set or other paradoxical sets (see
[46], [52]), [57]). To be of interest the underlying negation must be strong enough
to warrant that such sets are really paradoxical and not just only apparently or
poetical contradictory sets. Paraconsistent logic has not yet appeared as useful
for some new foundations of mathematics but the future will maybe bring some
surprises through some more audacious tentatives such as logics in which a
formula can be identical to its negation (cf. [67]).
3 Relations between paraconsistent logic and other
logics
It seems to us that the future of paraconsistent logic will depend on its relations
with other logics and in particular we dont share the belief of paraconsitent
Darwinians who have an evolutionary view of logic according to which only
paraconsistent logic will survive in the future because it is the strongest and best
logic. Neither we share the view of classical Darwinians and relevant creationists.
We think that logic is in an endless motion, that there is no nal super-logic,
that the nature of this motion can be studied within a general theory of logics
(cf. Universal logic), that a logic can properly be understood only by means of
a comparative study investigating its relationships with other logics.
8
3.1 Fuzzy logic and many-valued logic
Let us try to explain the relations between logics in terms of colours: classical
bivalent logic is black and white, a many-valued logic is so much multicoloured as
it is many-valued, fuzzy-logic is densely multicoloured including all the colours
between purple and pink, yellow and green, black and white. And so what
would be the colour of paraconsistent logic? Red, like the re of a revolutianory
logic? Deep blue, like the articial intelligence of the computers star of the third
millenium? Unfortunately it seems that its colour is not so poetical, maybe just
black and white with a pinch of white in the black and a pinch of black in the
white, like the Taos symbol. Of course it is possible to mix paraconsistent logic
with multicoloured many-valued logic and to get a pinch of red into a sea of
blue and a pinch of yellow into a black see.
From the technical point of view, the relationships are more prosaic. One
can for example use many-valued matrices to try to dene a paraconsistent
negation, but the success is not guaranteed. In the course of history, Asenjo
[5], DOttaviano/da Costa [58] and Priest [76] have presented paraconsistent
negations constructed with many-valued matrices, but none of these logics has
a satisfactory behaviour. In particular they all are non self-extensional (cf.
[26]). Many-valued matrices have also been used as a semantics for atomical
paraconsistent logics like Settes logic [84] and Puga/da Costas formulation of
Vasilievs logic [80]. More recently people like Karpenko [64], Avron [6], Tuziak
[92] or Almeida [2] have investigated in a more general way what can be done
with many-valued matrices in the eld of paraconsistent logic. Furthermore
the present author [16] has proposed to extend the traditional conception of
many-valuedness based on matrices to non truth-functional semantics and has
presented such semantics for the paraconsistent logic C1 and other ones.
At rst sight, many-valuedness seems an attractive idea for paraconsistent
semantics. But it may be just an illusion. One has to be careful with the mixture
between a binary division in the language (a and a) and multiple values in the
semantics, and more generally with the bendy philosophy of many-valuedness,
according to which there are in some sense many-values, but in another sense the
principle of bivalence is preserved with the binary division between designated
and non-designated values (see [51], [25]). Many-valued matrices are for sure a
useful mathematical tool for paraconsistent logic as it is for many logics such as
classical logic, when it is used for example to prove independence of axioms (cf.
[3]). But it is not so clear that they are philosophically relevant for paraconsis-
tency. History has shown that in the case of modal logic, many-valued matrices
do not work, although at rst sight it seems that there is a connection between
possibility and a third value. Maybe the same is true of paraconsistent logic in
the sense that a good paraconsistent logic cannot be characterized by a nite
matrix. Maybe, like for modal logics, Kripke semantics is better adapted (see
Section 3.3), or maybe the perfect paraconsistent semantics must be constructed
using new technics such as Carniellis possible-translation semantics (cf. [40],
9
[2]) or Buchsbaums min-max semantics [37].
3.2 Relevant logic
For many relevantists classical logic is totally wrong and must be rejected be-
cause it is not relevant, in the sense that you can deduce from a set of assump-
tions a conclusion which says something which has nothing to do with what the
assumptions are about. The ex-falso sequitur quodlibet is not a relevant de-
duction. For the relevantists this is one more example which shows how much
classical logic is wrong and they think that paraconsistentists are right to reject
it, but they also think that they must not stop there, they must go on and reject
all other non relevant deductions.
Relevantists like to say that paraconsistent logic is part of relevant logic
and paraconsistentists like to say the opposite. In fact a relevant logic can be
paraconsistent in no signicant way, and this is the case of the usual relevant
logics. This illustrates perfectly the fact that the rejection of the ex-falso se-
quitur quodlibet is not enough to dene paraconsistency. And it is also very
clear that it is possible to develop paraconsistent logics which are not relevant
: typical examples are paraconsistent logics which are nitely trivializable or
which admit a classical negation.
Maybe the relevantists are right and in the future our reasonings will be
completely meaningfully mechanized in such a way that it will not be anymore
possible to take a rabbit out of a hat, the magic of the ex-falso fading away. But
this does not necessarily imply that we will live in a world of contradictions. Or
maybe the paraconsistentists are right, we will live in a world full of meaningful
non exploding contradictions, but allowing also some magic touches of purely
formal deduction. Or maybe the logic of the future will be the daughter of
relevant logic and paraconsistent logic, a logic of contradictory meanings.
3.3 Modal and intensional logics
In the same way that we can extend classical logic into various modal logics (S4,
S5, B52, F242, etc) by adding the modal operators of possibility and necessity,
we can extend the various modal logics and neo-modal logics (tense, epistemic,
doxatic, deontic, faith, belief logics, etc) into paraconsistent logics by adding
paraconsistent negations.
There are many intuitive motivations, stronger in this context than in pure
extensional logic. For example, contradictory beliefs is a natural concept in our
world in which rockets are sent out of the solar system diusing music to make
the universe beat at the rhythm of human craziness, and at the same time many
people still think that earth is as at as a CD manufactured by a ingenious god
or a malin genie to play the eternal song of human sorrow.
But to deal with contradictory beliefs (fears, obligations, etc.) do we really
need a paraconsistent logic? What we want is to be allowed to believe in God
10
and Satan without admitting anything such as the existence of green creatures
on Mars. But maybe this can be done without rejecting the ex-falso sequitur
quodlibet. To believe in Satan does not necessarily mean not to believe in God,
it can mean to believe in a anti-God. Formally speaking (we use the symbol
for belief representation), we may have a logic of belief in which a, a b
but in which a, a b and more generally a, a b. So this logic may
not be, strictly speaking, a paraconsistent logic. The question to know if we
can deal with contradictory beliefs, in this sense, without entering the realm
of paraconsistency has been put forward by J.Wainer [98]. Anyway, in case of
a positive answer to this question, it seems to us that instead of saying that
there are logics of contradictory beliefs which are not paraconsistent, it would
be better to modify the denition of paraconsistency in order to include this
kind of logics within the realm of paraconsistency. More generally, given an
intensional operator , one can say that a logic in which a, a b but
in which the ex-falso holds is a paraconsistent logic, or to be more precise is a
heartian paraconsistent logic.
In fact, combinations of intensional operators such as standard modalities
with standard extensional operators such as classical negation can lead to para-
consistent negations. That what has been recently pointed out by the author
[30]. For example, in S5, is a paraconsistent negation with interesting prop-
erties. From this point of view, the relationships between modal logics and
paraconsistent logics seem to be a very promising eld of investigation for the
future: on the one hand we can produce interesting new paraconsistent negations
using intuitions and techniques of modal logics (in particular Kripke models),
on the other hand we can generate intensional operators with paraconsistent
negations.
3.4 Intuitionistic logic, paracomplete logics and paranor-
mal logics
Intuitionistic logic appears as a dual of a particular paraconsistent logic. Re-
verse intuitonistic logic, put his head down and his foots up, his foots will look
like a head and his head like some foots, and you will get another logic, a para-
consistent logic. But there are several ways to reverse intuitonistic logic, it all
depends on how you see it, you dene it. If you look at it through a sequent
calculus, you will reverse this vision by admitting only one formula on the left
and you will get something [95] which is not necessarily the same as the reverse
vision of another vision, such as a the dual of a Heyting algebra [85], the dual
of a Kripke semantics, or the dual of a topos (cf. [66], [63], [97]).
Moreover there are many other paraconsistent logics than all the reverse
visions of intuitionistic logic. And all the reverses of paraconsistent logic are
many more than all the possible visions of intuitionistic logic, and form the
rich eld of paracomplete logics [68]. Each paraconsistent logic has a paracom-
plete dual and each paracomplete logic has a paraconsistent dual. The duality
11
between paraconsistency and paracompleteness can be worked out with any se-
mantical idea, informal interpretation, etc. For example from the viewpoint of
game semantics, following the idea of T.Pequeno, paraconsistency corresponds
to a game where both players can win, by contrast to paracompletness which
corresponds to a game where both players can loose. Paraconsistent logic and
paracomplete logic appear therefore like husband and wife. The insidious anity
between paraconsistency and paracompletness is expressed in North-American
short and imaginative way of speaking by the expression Gaps and Gluts.
There are also logics with gaps and gluts. Leaving aside comic strip termi-
nology and using again a neo-classical para terminology, such logics, which
are both paraconsistent and paracomplete, are called paranormal logics or some-
times non-alethic logics (Miro Quesadas terminology). Among paranormal log-
ics, we nd De Morgans logics (see [13], [60]). Paranormal logics may have
interesting applications to intensional logic or articial intelligence. (on para-
normal logics see e.g. [15], [61]).
Paraconsistency is an open door to the promising elds of paracomplete
and paranormal logics. Therefore, even if in the future it is paracomplete and
paranormal logics that will dominate the logical world, paraconsistency will play
an essential role in this domination.
3.5 Non monotonic logic, alfabar logic and substructural
logics
After rejecting the principles of excluded middle and contradiction, one can
wonder what else can be dropped, if we are not already in presence of a naked
meaningless conception of logic. But these last years the logicians have gone
farther than post-modern artists, they have pursued the deconstruction up to
the structural level and study logics called substructural logics. The relations
between paraconsistent logic and substructural logics may be of various kinds.
Paraconsistency can appear as an alternative, replacing the rejection of struc-
tural principles, with the rejection of logical principles, escaping the substruc-
tural world, or as a complement, crowding the substructural world with contra-
dictions.
Paraconsistent logic can be seen as an alternative, for example, to non-
monotonic logic. Non-monotonists reject monotony because they think that
there are experiences (most of the time involving birds) which show that monotony
is wrong and in particular leads to some contradictions. But one who thinks
the paraconsistent way would reject the principle of non contradiction and not
monotony. The strategy of the paraconsistentist is more imaginative, he accepts
to see penguins ying in the sky of Hawais beaches and pink oyds surng on
Antarcticas permafrost. It seems to us that the future shall give the preference
to paraconsistent logic taking in account the progress of genitical biology which
already produces chicken without feathers, and in the future we may have y-
ing pigs. In such an absurd world, it will make no sense to reason by default,
12
because everything could be true by default.
Paraconsistent logic can also be seen as a complement, for example, to al-
fabar logics. These are logics in which the law of autodeductibility, a a, is
not in general valid (cf. [65]). One can dene a paraconsistent negation over
such substructural logics. In such logics it may happen that a, a a. Mixing
paraconsistency with alfabaricy may be a good solution to get neutral paracon-
sistent logics, paraconsistent logics in which from a contradiction it is impossible
to deduce anything, in exact contrast to classical logic in which a contradiction
produces an deductive explosion.
3.6 Non Frego-Aristotelean logic
Consider that the planet LOGOS of the logicians is a rectangle with a length
of 100 km and that on the left extermity of the planet (right from the point of
view of some Extra-Logosians) there is a big group of serious guys whose god
is classical logic they venerate six days a week and that on the right extremity
there is a not less numerous crowd of much more agitated people, some of them
running in circle trying to catch their own tail, some others trying to jump
outside of LOGOS in the middle of nothing. Paraconsistentists seem rather
close to the right board of the planet LOGOS. But even if the paraconsistentist
is the crazy guy, at the extreme right of the LOGOS planet, trying to jump
to another planet, nonetheless he is still on the same logical plane than the
classical logician. Both are grounded on some similar Fregean and Aristotelean
onto-logical dogmas.
These dogmas are mutliple and form a conception of logic which is still
predominant nowadays but seems to be ready to explode. For example one
central feature of Frego-Aristotelean logic, its formal aspect, is now seriously
attacked. It is challenged by people putting meaning into logic but also by people
putting meaning onto logic, showing that it is possible to rigorously reason with
diagrams such as Venns diagrams (see [86]). For sure the true revolution of
the third millenium logic will be rather connected to a visual logic than to a
logic that one gets from classical logic by toying with standard formalizations,
dropping one axiom, slightly modifying one other, deleting the left structural
rule of right contraction, adding truth values, etc, giving birth to monstrous
creatures with short life expectancy.
If paraconsistent logic wants to survive the 20th century it should already
irt with non Frego-Aristotelean logics. An interesting easy project, for exam-
ple, will be to show how to reason with paraconsistent Venns diagrams. Maybe
such paraconsistent diagrams can bring new ideas into the visual anlysis of logic
in such a way that paraconsistent logic will not be in the future just an accessory
tool but part of the kernel of the new sphere of rationality.
13
4 Applications of paraconsistent logic
A logic can grow like a majestic tree, dominating the logical orest by its beauty
and grandiosity, but if such a tree produces no fruits, it can turn into a monstrous
cadaver of black wood soon to disappear.
Like for many non classical logics, the future of paraconsitent logic will in
great part depends on its possible applications. Let us therefore examine if
paraconsistent logic is a fruitful tree, or a sterile plant that must be removed
from the garden of science.
4.1 Applications to human sciences
Maybe human being is the only natural phenomenon which is contradictory,
producing contradictions and seeing them everywhere. Obviously human being
is full of contradictions. The question is to know if they must be bannished and
rejected considering them as errors and source of confusions, or if they must be
taken as natural, an essential part of human nature we must deal with.
Given a human being like John Smith, with contradictory desires and wills,
can we think that it is his normal state, that John Smith behaves in a paracon-
sistent way and that paraconsistent logic is the adequate tool to describe his
behaviour? Or must we think that these contradictions are a kind of disease
that should be eliminated, for John Smith recovering his health, following again
the pattern of classical logic?
Are contradictions irrational? Is paraconsistent logic the logic of parapsy-
chology, lacanian inconscious, sokalian impostures, cyber yuppies? Is paracon-
sitent logic just a curiosity which ts well the spirit of the post-modern bric-
a-brac of new age rationalism? Or is paraconsistent logic the logic of political
correctness, of minorities, pluralities, the logic of oecumenical concilianility of
the Aquarian era?
It is dicult to answer all these questions. For the present time, there is
on the one hand an odd mathematical tool and on the other hand apparently
contradictoy phenomena, described with languages which have structures very
dierent to mathematics.
We must investigate, little by little, applications of paraconsistent logic and
mathematical modelizations of human phenomena to see if we nd a convergent
point. One can start by working with more concrete cases, such as for example
the case of lie dectectors and contradictory witnesses [27].
4.2 Applications to natural sciences
Are there contradictions in the nature ? Let us rst note that someone who
believes in real contradictions does not necessarily believe in natural contradic-
tions, since for example linguistic contradictions may be considered as real but
not part of the nature. It all depends of course of how we dene and conceive
14
nature. But in general linguistic is considered as a human science by opposition
to natural sciences such as physics and biology.
Contradictions can appear in many dierent ways within physics. Some peo-
ple think that the physical world is bound by the interelationship of antagonistic
forces. Anyway the underlying standard logic of physics is classical logic. This
does not mean that another logic based on contradictions could not lead to an
equivalent theory or a better one, but until now such kind of possibility has
not been developed in a way as systematic and general as the standard frame-
work. Even in the case of quantum physics, where contradictions appear in a
phenomenon such as the duality wave/particle.
The people of the Copenhagues school have developed an intepretation of
this duality based on the notion of complementarity. Such an interpretation can
be seen as having a touch of paraconsistency. But Copenhagues interpretation
was presented rather in a philosophical informal way by Bohr and Heisenberg.
Paraconsistency could be used as a possible formalization of it, as alternative
to other previous formalizations such as the many-valued logic of P.Fevrier [59],
a logician who took seriously the idea of complementarity. People like N.C.A.
da Costa and D.Krause are presently working on a paraconsistent approach to
complementarity [55]. Another Costas idea involving paraconsistent logic into
the formalization of physics, is multideductive logic, a way to unify contradictory
theories, worked out in details by E.G. de Souza [88]. In the future, if nobody
succeeds to unify micro and macro physics into a classical theory, maybe
paraconsistent logic will be an essential tool for the foundations of physics.
Biological phenomena can even more easily be interpreted as contradic-
tory phenomena . Hegel used to take the growth of a plant as a typical ex-
ample of contradictory natural phenomenon. The question it know how this
contradiction-oriented approach can be systematized and fromalized. Until now,
no serious mathematical system based on contradictions has been proposed for
biology. But it is also true that in general, applications of logic to biology are
quite rare despite of the interest of Aristotle and Tarski for this science. Maybe
it is because classical logic does not t at all biological phenomena. Maybe
then in the future, a logic like paraconsistent logic will be useful to turn biology
logical.
4.3 Applications to formal sciences
By formal sciences, we understand here logic, mathematics and any other general
abstract nonsense.
When, a century ago, Cantor created a paradise full of alephes, and other
sensual creatures that seem to be only product of a endless imagination, he was
arguing against the theologians, that mathematics is free, in the sense that the
mathematician can entertain himself with any kind of creatures, unless there
are contradictory. Contradictions are the end of the game. Contradictions are
like snakes in Cantors paradise. If you meet a snake, he will bite you and you
15
will have to go out of the paradise of alephes, awaking in everyday world, full
of politicians and pollution.
The mathematician is therefore in some sense totally free, he can create
anything even if these things have no applications or are not in any sense rep-
resentations of reality. But in another sense he is not so free : he is severely
bound by the principle of non contradiction. He cannot, like the poet, live in a
world of ying rabbits, happy alephes and true liars.
Maybe the paraconsistentist wants to turn mathematics more poetical, al-
lowing Russellian sets and other funny creatures in Cantors paradise, such as
friendly snakes.
However it seems that paraconsistent logic does not broaden signicantly
enough the mathematical horizon. If we replace consistency with non triviality,
we get a mathematical world which is much the same as the classical world.
It is not clear to which point this paradigmatical change permits to introduce
new mathematical concepts of interest. New creaures, like a Russellian set, are
maybe only funny ugly ornaments added to the beautiful classical architecture
of mathematics, as the gargoyles added to the majestic architecture of Notre
Dame.
But maybe these new creatures authorized in the mathematical paradise by
paraconsistent logic will play a central role in the future scene of mathematics,
maybe paraconsistent logic will give, at least, denitive citizenship to innitesi-
mals in the mathematical country (cf. [72]). Or maybe paraconsistent logic will
save us from the tricephalous CGC-monster (CGC for Cantor-Godel-Church)
by providing foundations for nite decidable complete mathematics (see [14]).
It is too soon to have an opinion, but against the skeptics who see these crea-
tures as disturbing parasites, toys for infantil logicians, we can recall George
Birkhos remark : It is well not to forget that many of the most astonishing
mathematical developments began as a pure jeu desprit.
4.4 Applications to cyber sciences
The expression cybernetics is a word from the prehistory of our modern com-
puter times that has started to be used again in the eighties in a shorter design
(cf. cyber space). Following Prof. Castigo, from the Santa Boneca Institute,
we will use it in its shorter design, as a generic term for information sciences,
cognitive science, articial intelligence, computational intelligence, etc, all these
sciences converging towards the birth of a super robot, a cyber robot.
For sure cyber science will rule the future in a way that we can imagine only
supercially. Let us see if paraconsistent logic will play a signicant role in this
future or if we shall be regulated by classical bivalent robots not even able to
laugh at a contradiction.
The Prof. Tsujiski just informed us about a recent experience that has been
performed in the Butatin Studio of the Center for Avanced Study of the Univer-
sity of Sao Paulo, under the guidance of Prof. Abe [1]. A paraconsistent female
16
robot called Sophia and her brother Anti-Sophia, have survived a confrontation
with tigers, snakes and monkeys using the paraconsistent logic petale which
rules the electronic circuits of their neural networks (But Prof. Tsujiski said
nothing to us about the ability of these paraconsistent robots to survive the
stress of everyday life in Wall Street).
In our opinion this experience, as well as other similar experiences (cf. [56]),
show very well the future of paraconsistent logic: We dont know if paraconsis-
tent logic is just a crazy dream of our present times that will soon disappear
or the new logic of a future wonderful more colourful world, but fore sure to-
morrow, even in a not so wonderful world, there will be a cohort of intelligent
paraconsistent robots ruling our world, maybe with many fuzzy brother robots,
and many non monotonic sister robots. However we cannot predict if their big
brother will be a pure paraconsistent robot, or a substructural paraconsistent
turbo polar robot [75]).
Maybe in the future paraconsistent logic will go out of the logico-mathematical
scene, manifesting itself in an important way, only in the cyber engineer scene.
5 Paraconsistent logic and the third millenium
Some paraconsistent new agers like to say, parodying Malraux, that the third
millenium will be paraconsistent or will not be. We dont know exactly what
they mean. For sure, if we succeed to escape big-bug annihilation, it will not be
via paraconsistent logic. The big-bug problem is probably full of contradictions,
but with January 1st 2000 approaching at a speed of 2160 seconds per hour it is
very unprobable that paraconsistent logic can help in our salvation, since it has
not even yet been applied to bug problems. So if we enter the third millenium
this will be thanks to old black and white binary classical logic. But of course
this does not mean that classical logic will be able to guide us through the whole
third millenium. Maybe paraconsistent logic will take the relay and help us to
cope with the next big-bang.
References
[1] J.M.Abe, Some recent applications of paraconsistent systems to AI,
Logique et Analyse, 157, 83-96.
[2] J.M. de Almeida, 1999, Possible translations semantics, Master Thesis,
Cle-Unicamp.
[3] E.H.Alves, 1976, Logic and inconsistency : study of the calculi Cn, Master
Thesis, Cle-Unicamp.
[4] O.Arieli and A.Avron, 1996, Reasoning with logical bilattices, Journal
of Logic, Language and Information, 5, 25-63.
17
[5] F.G.Asenjo, 1966, A calculus of antinomies, Notre Dame Journal of For-
mal Logic, 7, 103-105.
[6] A.Avron, 1990, Relevance and paraconsistency - a new approach, Journal
of Symbolic Logic, 55, 707-732.
[7] A.I.Arruda, 1984, N.A.Vasiliev : a forerunner of paraconsistent logic,
Philosophia Naturalis, 21, 472-491.
[8] D.Batens, 1980, Paraconsistent extensional propositional logics, Logique
et Analyse, 90-91, 195-234.
[9] D.Batens, 1994, Inconsistency-adaptive logics and the foundation of non-
monotonic logic, Logique et Analyse, 145, 57-94.
[10] D.Batens, 1997, Inconsistencies and beyond. A logical-philosophical dis-
cussion, Revue Internationale de Philosophie, 200, 259-273.
[11] D.Batens et al. (eds), Frontiers in paraconsistent logic, to appear.
[12] V.A.Bazhanov, 1990, The fate of a forgotten idea: N.A.Vasiliev and his
imaginary logic, Studies in Soviet Thought , 37, 143-151.
[13] N.Belnap, 1977, A useful four-valued logic, in Modern uses of multiple-
valued logic, J.M.Dunn and G.Epstein (eds), Reidel, Dordrecht, pp.8-37.
[14] J.P. van Bendegem, The strong Hilbert program, to appear.
[15] J.-Y.Beziau, 1989, Calcul des sequents pour logique non-al`ethique,
Logique et Analyse, 125-126, 143-155.
[16] J.-Y.Beziau, 1990, Logiques construites suivant les methodes de da
Costa, Logique et Analyse, 131-132, 259-272.
[17] J.-Y.Beziau, 1993, Nouveaux resultats et nouveau regard sur la logique
paraconsistante C1, Logique et Analyse, 141-142, 45-58.
[18] J.-Y.Beziau, 1994, Universal logic, in Logica94 - Proceedings of the 8th
International Symposium, T.Childers and O.Majer (eds), Czech Academy
of Sciences, Prague, pp.73-93.
[19] J.-Y.Beziau, 1994, Theorie legislative de la negation pure, Logique et
Analyse, 147-148, 209-225.
[20] J.-Y.Beziau, 1995, Negation : what it is and what it is not, Boletim da
Sociedade Paranaense de Matematica, 15, 37-43.
[21] J.-Y.Beziau, 1995, Recherches sur la logique universelle, PhD, Univeristy
of Paris 7, Paris.
18
[22] J.-Y.Beziau, 1996, On logical truth, PhD, University of S ao Paulo, Sao
Paulo.
[23] J.-Y.Beziau, 1996, Identity, logic and structure, Bulletin of the Section
of Logic, 25, 89-94.
[24] J.-Y.Beziau, 1997, Logic may be simple, Logic and Logical Philosophy,
5, 129-147.
[25] J.-Y.Beziau, 1997, What is many-valued logic ?, in Proceedings of the
27th International Sympoisum on Multiple-Valued Logic, IEEE Computer
Society, Los Alamitos, pp.117-121.
[26] J.-Y.Beziau, 1998, Idempotent full paraconsistent negations are not alge-
braizable, Notre Dame Journal of Formal Logic, 39, 135-139.
[27] J.-Y.Beziau, 1998, Applications de la logique paraconsistante la justice et
au droit, in Anais do V Congresso Brasileiro de Filosoa, vol.2 , Brazilian
Philosophical Institute, Sao Paulo, 1119-1128.
[28] J.-Y.Beziau, 1999, Classical negation can be expressed by one of its
halves, Logic Journal of the Interest Group on Pure and Applied Logics,
7, 145-151.
[29] J.-Y.Beziau, Paraconsistent logic !, to appear.
[30] J.-Y.Beziau, S5 is a paraconsistent logic and so is rst-order classical
logic, to appear.
[31] J.-Y.Beziau, From paraconsistent logic to universal logic, to appear in
Synthese.
[32] J.-Y.Beziau, The paraconsistent logic Z - A possible solution to Jaskowkis
problem, to appear.
[33] J.-Y.Beziau, Complexical paraconsistent logic, to appear.
[34] J.-Y.Beziau, What is paraconsistent logic?, to appear in [11].
[35] A.Bobenrieth, 1996, Inconsistencias, Por que no?, Colcultura, Bogota.
[36] A.Bobrihna do Lar, 1761, Estudos asistematicos da lingua Puppy-Guarana,
Curitiba.
[37] A.Buchsbaum, 1995, Logics of inconsistency and incompleteness: seman-
tics and axiomatization, Catholic University of Rio de Janeiro, Rio de
Janeiro.
[38] W.Bychovski, 1987, Polar paraconsistent logic, Proceedings of the Polda-
vian Academy of Sciences, 37, 123-156.
19
[39] C.E.Caorsi, 1994, L ogica, losoa y psicoan alisis, Roca Viva, Montevideo.
[40] W.A.Carnielli, Possible-translation semantics for paraconistent logic, to
appear in [11].
[41] N.C.A. da Costa, 1963, Calculs propositionnels pour les syst`emes formels
inconsistants, Comptes Rendus de lAcademie des Sciences de Paris, 257,
3790-3793.
[42] N.C.A. da Costa, 1964, Calculs des predicats pour les syst`emes formels
inconsistants, Comptes Rendus de lAcademie des Sciences de Paris, 258,
27-29.
[43] N.C.A. da Costa, 1992, La losoa de la logica de Francisco Miro Quesada
Cantuarias, in Logica, Razon y Humanismo, Lima, pp.69-78.
[44] N.C.A. da Costa, 1997, Logiques classiques et non classiques , Masson,
Paris.
[45] N.C.A. da Costa and J.-Y.Beziau, 1994, Theorie de la valuation, Logique
et Analyse, 146, 95-117.
[46] N.C.A. da Costa and J.-Y.Beziau, 1996, Theorie paraconsistante des en-
sembles, Logique et Analyse, 153-154, 51-67.
[47] N.C.A. da Costa and J.-Y.Beziau, 1997, Overclassical logic, Logique et
Analyse, 157, 31-44.
[48] N.C.A. da Costa and J.-Y.Beziau, 1997, Denitions, theorie des objects
et paraconsistance, Theoria, 32, 367-379.
[49] N.C.A. da Costa, J.-Y.Beziau and O.A.S.Bueno, 1995, Paraconsistent
logic in a historical perspective, Logique et Analyse, 150-151-152, 111-
125.
[50] N.C.A. da Costa, J.-Y.Beziau and O.A.S.Bueno, 1995, Aspects of para-
consistent logic, Bulletin of the Interest Group in Pure and Applied Logics,
4, 597-614.
[51] N.C.A. da Costa, J.-Y.Beziau and O.A.S.Bueno, 1996, Malinowski and
Suszko on many-valuedness: on the reduction of many-valuedness to two-
valuedness, Modern Logic, 6, 272-299.
[52] N.C.A. da Costa, J.-Y.Beziau and O.A.S.Bueno, 1998, Elementos de teoria
paraconsistente de conjuntos, Cle-Unicamp, Campinas.
[53] N.C.A. da Costa and M.Guillaume, 1964, Sur les calucls Cn, Anais da
Academia Brasileira da Ciencias, 36, 379-382.
20
[54] N.C.A. da Costa and M.Guillaume, 1965, Negations composees et loi de
Peirce dans les syst`emes Cn, Portugalia Mathematicae, 24, 201-210.
[55] N.C.A. da Costa and D.Krause, Paraconsistency and complementarity,
to appear.
[56] N.C.A. da Costa and V.S.Subrahmanian, 1989, Paraconsistent logics as
a formalism for reasoning about inconsistent knowledge bases, Articial
Intelligence in Medecine, 1, 167-184.
[57] R. da Costa Caiero and E.G. de Souza, A new paraconsistent set theory:
ML1 , Logique et Analyse, 157, 115-141.
[58] I.M.L.DOttaviano and N.C.A. da Costa, Sur un probl`eme de Jaskowski ,
Comptes Rendus de lAcademie des Sciences de Paris, 270A, 1349-1353.
[59] P.Fevrier, 1937, Les relations dincertitude dHeisenberg et la logique,
Travaux du IXeme Congr`es Internationalde Philosophie, Tom. VI, Her-
mann, Paris.
[60] J.M.Font, 1997, Belnaps four-valued logic and De Morgan lattices, Logic
Journal of the Interest Group in Pure and Apllied Logic, 5, 413-440.
[61] N.Grana, 1990, Contradizione e Incompletezza, Liguore, Napoli.
[62] S.Jaskwoski, 1948, Rachunek zda n dla systemow dedukcyjnych
sprzecznych, Studia Societatis Scientiarum Toruniensis , 1A, 57-77.
[63] W.James and C.Mortensen, Categories, sheaves and paraconsistent logic to
appear,
[64] A.Karpenko, 1986, Paraconsistent structure inside of many-valued logic,
Synthese, 66, 63-69.
[65] D.Krause and J.-Y.Beziau, 1997, Relativizations of the principle of iden-
tity, Logic Journal of the Interest Group in Pure and Apllied Logic, 5,
327-338.
[66] F.W.Lawvere, 1991, Intrinsic co-Heyting boundaries and the Leibniz rule
in certain toposes, in Category theory, Springer, Berlin, pp.279-281.
[67] L.Lismont, 1992, Logique du premier ordre anti-fondee, Cahiers du Cen-
tre de Logique, 7, 69-95.
[68] A.Loparic and N.C.A. da Costa, 1984, Paraconsistency, paracompleteness
and valuations, Logique et Analyse, 106, 119-131.
[69] A.T.C.Martins, 1997, A syntactical and semantical uniform treatement for
the IDL and LEI nonmonotonic systems , PhD, Federal University of Per-
nambuco, Recife.
21
[70] C.Mortensen, 1980, Every quotient algebra for C1 is trivial, Notre Dame
Journal of Formal Logic, 21, 694-700.
[71] C.Mortensen, 1989, Paraconsistency and C1, in [79], pp.289-305.
[72] C.Mortensen, 1995, Inconsistent Mathematics, Kluwer, Dordrecht.
[73] T.Pequeno and A.Buchsbaum, 1991, The logic of epistemic inconsistency,
in Principles of knowledge representation and reasoning , Morgan Kauf-
mann, San Mateo, pp.453-460.
[74] L.Pena, 1979, Contradiction et verite, PhD, Univeristy of Li`ege, Li`ege.
[75] J.Pipoca del Mar, 1999, Substructural paraconsistent turbo polar robots :
how they behave and why they will rule our future, Padernos do Instituto
do Planejamente Central da Alta Bagunca, 678, 2221-3098.
[76] G.Priest, 1979, Logic of paradox, Journal of Philosophical Logic, 8, 219-
241.
[77] G.Priest, 1987, In contradiction. A study of the transconsistent , Nijho,
Dordrecht.
[78] G.Priest, Paraconsistent logic, to appear in Handbook of philosophical
logic, Second Edition, D.Gabbay et al. (eds), Kluwer, Dordrecht.
[79] G.Priest, R.Routley and J.Norman (eds), 1989, Paraconsistent logic : Es-
says on the inconsistent , Philosophia, Munich.
[80] L.Puga and N.C.A. da Costa, 1988, On the imaginary logic of
N.A.Vasiliev, Zeitschrift f ur mathematische Logik und Grundlagen der
Mathematik, 34, 205-211.
[81] A.P.Pynko, 1995, On Priests logic of paradox, Journal of Applied Non-
Classical Logics, 5, 219-225.
[82] W.V.O.Quine, 1970, Philosophy of logic, Englewood Clis.
[83] R.Routley and V.Routley, 1985, Negation and contradiction, Revista
Colombiana de Matem aticas, 19, 201-231.
[84] A.M.Sette, 1973, On the propositional calculus P1, Mathematica
Japonae, 16, 173-180.
[85] A.M.Sette, E.H.Alves and G.S. de Queiroz, Brouwerian algebras and para-
consistent logic, to appear.
[86] Sun-Joo Shin, 1994, The logical status of diagrams, Cambridge University
Press, Cambridge.
22
[87] B.H.Slater, 1995, Paraconsistent logics ?, Journal of Philosophical Logic,
24, 451-454.
[88] E.G. de Souza, 1995, Destouchess problem and heterodoxical logics : essay
on the use of non classical logics in the treatement of inconsistencies in
physics, PhD, University of Sao Paulo, S ao Paulo.
[89] R.Sylvan, 1990, Variations on da Costa C-systems and dual-intuitionistic
logics. 1. Analyses of C

and CC

, Studia Logica, 49, 47-65.

[90] R.Sylvan and I.Urbas, 1993, Paraconsistent classical logic, Logique et
Analyse, 141-142, 3-24.
[91] M.Tsuji, A paraconsistent theory of decision under uncertainty, Logique
et Analyse, 157, 101-114.
[92] R.Tuziak, 1997, Finitely many-valued paraconsistent systems, Logic and
Logical Philosophy, 5, 121-127.
[93] I.Urbas, 1987, On Brazilian paraconsistent logics, PhD, Canberra, 1987.
[94] I.Urbas, 1990, Paraconsistency, Studies in Soviet Thought , 39, 343-354.
[95] I.Urbas, 1996, Dual-intuitionistic logic, Notre Dame Journal of Formal
Logic, 37, 440-451.
[96] M.Urchs, 1995, Discursive logic. Towards a logic of rational discourse,
Studia Logica, 54, 231-249.
[97] V.L.Vasyukov, 1993, A categorical semantics for paraconsistent logic, in
Logical investigations, vol.2, Moscow, pp.285-298.
[98] J.Wainer, How to join two sources of information, to appear.
[99] R.Wojcicki, 1988, Theory of logical calculi , Reidel, Dordrecht.
[100] G.H. von Wright, 1986, Truth, negation and contradiction, Synthese,
66, 3-14.
[101] G.H. von Wright, 1987, Truth logics, Logique et Analyse, 120, 311-334.
Jean-Yves Beziau
Universidade Federal Fluminense
Instituto de Matematica
Departamento de Analise
Niteroi, Rio de Janeiro, Brazil
[email protected]
23