II
METAPHILOSOPHY AS CONCEPTUAL ANALYSIS
Eine ganze Wolke von Philosophie kondensiert zu einem Tröpfchen Sprachlehre.
[A whole cloud of philosophy condenses into a drop of grammar.]
Wittgenstein
As descriptive metaphilosophers, when we look back at the history of
philosophy, we often encounter explanations of its nature that we feel inclined
to dismiss outright. One example is the attempt to ground philosophy in a
single object or a unique method. After all, there is nearly as much variety in
objects and methods as there is in philosophies and philosophical movements themselves.[1]
Moreover, the many subdivisions
of theoretical philosophy – as concerned with the world’s input upon the
subject – and practical philosophy – as concerned with the subject’s output
upon the world – seem to correspond to an equally wide range of specific objects
and with methodologies that vary according to the demands of each subddivision.
Only the prescriptivist
metaphilosopher may still entertain the hope, or the illusion, of defining philosophy’s
proper object of inquiry. The descriptivist metaphilosopher, by contrast, will tend
to regard such attempts as inherently reductionist or expansionist, for they
either narrow or broaden philosophy’s boundaries beyond what would be most appropriate.
Because my aim is more constructive
than critical, I’ll focus on exploring some central analytical approaches in philosophy.
This idea shaped some of the most important developments in early 20th‑century
thought, especially the hugely influential view that philosophy’s proper method
is conceptual analysis, and its proper subject is the structure of
concepts—or, as Ernst Tugendhat put it, the framework of concepts most
central to how we understand the world.[2]
This conception has been
supported in various ways by philosophers such as Wittgenstein, Bertrand
Russell, G. E. Moore, Friedrich Waismann, A. J. Ayer, P. F. Strawson, Michael
Dummett, Ernst Tugendhat, and R. E. Brandom, among many others. Furthermore, as
we shall see, it can be extended to the philosophical tradition if we recall Socrates
in Plato’s dialogues – whose recurring question took the form “What is X?”,
where X stood for conceptual terms such as ‘friendship’, ‘love’, ‘beauty’, ‘justice’,
‘knowledge’, etc. The answer of questions of the form “Wha tis X?” should be a definition
of the concept achieved in Plato’s dialogues by division or dihairesis
(διαίρεσις), in other words, by a kind of analysis of those concepts.
Nonetheless,
the conception of philosophy as pure conceptual analysis was seriously
challenged by the so‑called “naturalistic turn,” initially advanced by W. V. O.
Quine from a deflationary perspective.[3] For
him, philosophy is not reducible to a purely linguistic‑conceptual inquiry,
since it is not essentially distinct from empirical science.
In fact, in his view there is
not even a real distinction to be drawn between the two: philosophy forms a
continuum with science, and the distinctions that can be drawn are merely
artificial, like the borders between the different states of the same country.[4]
Whatever advantages this point
of view may have, the problem remains that no defender of the claim that the
distinction between science and philosophy is merely arbitrary has ever been
able to explain why we are so reluctant to conceive of the boundary between the
two as the product of arbitrary agreement. The Quinean thesis – that the distinction
between philosophy and science is the result of an artificial decision – fails to
account for our strong resistance to the idea of shifting those boundaries, for
example, by reclassifying as science what has traditionally been called
philosophy, and vice versa. More importantly, and this seems to me decisive,
the thesis also fails to explain why we can recognize a new theory as either
philosophical or scientific without appealing to any conventional agreement.
The conception of philosophy
as conceptual analysis had, at least, the merit of attempting to answer these
questions by explaining the distinctive features of philosophical practice.
Furthermore, recognising that philosophy is distinguished from science by its
own means does not necessarily imply the admission of what Timothy Williamson
called exceptionalism – the idea that the methods of philosophy are
intrinsically unique, essentially distinct and, supposedly, superior to those
of science.[5]
Although there are a variety
of versions of the conception of philosophy as conceptual analysis, I will
reduce them here somewhat artificially to two general forms, in order to better
highlight the qualities and limitations intrinsic to this conception. I will
call these two variants:
A) CRITIC OF LANGUAGE
B) ANALYSIS OF LANGUAGE
Through the critique of language, we seek to analyze or elucidate
concepts with the aim of dissolving philosophical confusions. By contrast,
through the analysis of language, the goal is to examine concepts in
order to gain a deeper understanding of our conceptual architecture. In what
follows, I will give my account of each of these approaches to philosophy. I
will argue that, although both possess methodological merits, they fail to
provide an adequate explanation of philosophy’s true nature.
1. SHORTCUTS OF CRITIQUE OF LANGUAGE
The critique of language seeks to expose flaws in philosophical
arguments, many of them inherited from the tradition itself. Historically, this
endeavor has taken two distinct paths. The first lies in the application of
formal methods to the investigation of philosophical terms and propositions –
what I shall call formally oriented analysis. The second has a pragmatic
emphasis. It consists in carefully examining the meanings and uses of expressions
in our natural language as they arise within interpersonal contexts – what I
shall call communicationally oriented analysis.
I employ these expressions,
respectively, in place of a narrower distinction: the old contrast between the philosophy
of ideal language, guided by the mathematical logic of Frege and Russell,
and the philosophy of ordinary language, shaped by the everyday idiom as
in the later Wittgenstein and the Oxford philosophers of the mid‑twentieth
century.[6] The
reason is that the old distinction is somewhat limited in the presente days.
While it does acknowledge the divide between the philosophy of logical atomism
in Russell and in the Tractatus – as examples of a philosophy of ideal
language – and Wittgenstein’s therapeutic philosophy in the Philosophical
Investigations, together with the work of Oxford philosophers such as
Gilbert Ryle, J. L. Austin, and P. F. Strawson – as examples of ordinary
language philosophy – it fails to account for later developments within these
two traditions, one tending toward formalism, the other toward empiricism.
It overlooks, for instance, the
opposition between the heirs of ideal language philosophy who employed new
formal tools such as modal logic in Kripke, Putnam, and Kaplan, and the later,
more systematic heirs of ordinary language philosophy, such as John Searle with
his theory of intentionality, Paul Grice with his pragmatic maxims, and even
Jürgen Habermas with his theory of universal pragmatics.
We shall observe a curious difference
between formally oriented and communicationally oriented philosophy, one that
calls for careful explanation. The former presents stimulating metaphysical
challenges, though its plausibility is often in doubt; the latter offers
responses of greater intuitive appeal, though these may at times prove
insufficient or even trivial. Only in due course will we be able to explain the
reason of this difference.
The critique delivered by
formally oriented language can be illustrated through the investigations of
analytic philosophers, such as Bertrand Russell. One example helps to clarify:
a reason underlying Plato’s invention of his doctrine of Ideas may have been a confusion
arising from the superficial identity between the grammatical structures of
sentences such as “Beauty is pleasant” and “Socrates is bald.”
Induced by this identity, Plato
would have concluded that, since the subject of sentences such as the latter is
a proper name referring to a particular, the subject of sentences such as the
former must likewise be a proper name, and thus must refer to a particular,
which he named Beauty itself. However, since the “Beauty itself” cannot
be found in the visible world, it must reside in a realm accessible only to the
intellect: the transcendent world of Ideas.
Against this conclusion, the critique of language, based on modern
predicate logic, makes clear that the logical structures of the two types of
sentence are only apparently identical. In fact, the first sentence has a logical
structure quite different from its superficial grammatical form. Whereas “Socrates
(a) is bald (P)” has the logical form “Fa,” a sentence such as “Beauty is
pleasant” is logically analyzable as an abbreviation of “For every x, if x is
beautiful (Q), then x is pleasant (R),” or, simbolically: “(x)(Qx → Rx).” In
this analysis, the supposed proper name ‘Beauty’ disappears, while ‘is beautiful’
reveals itself as a mere predicative expression.
This distinction suggests that,
lacking modern predicate logic, Plato was misled by the superficial identity of
the superficial subject–predicate structure in both types of statements. Such
grammatical confusion provided an illusory justification for his formulation of
an ontology of transcendent Ideas. This was a metaphysical construction of
undeniable historical importance, yet one whose foundations, in light of
contemporary logical analysis, have become more questionable.
Let us now turn to an example
of critique directed at language in its communicative orientation, one closely
aligned with what Wittgenstein called “therapeutic philosophy.” It concerns the
exposition of philosophical confusions arising from the argument from illusion.
This argument was developed in opposition to direct or naïve
realism, which holds that we have immediate access to the external world. Its
purpose was to defend indirect realism (representationalism),
which holds that our access to the external world is mediated by sense-data,
or even to advance a counterintuitive phenomenalist idealism, which denies
access to any material world beyond the sense-data themselves.
This line of argument begins
with cases in which objects appear different from what they truly are, such as
the spoon that, when partially immersed in a glass of water, seems bent, or, in
more extreme cases, lifelike hallucinations that correspond to nothing in the
external world. The analysis of such cases seems to lead to the conclusion that
we perceive objects only indirectly: what we apprehend directly are not the
material things themselves, but rather our sensory impressions of them – the so-called
sense-data.
In opposition to such a conclusion, critics of
language, mainly J. L. Austin, have argued that we do not say we perceive only
representations rather than the objects themselves; what we actually say is
that we see the objects (like the spoon in the glass of water) directly,
though not as they truly are. Thus, when I look at my nose with both eyes, I do
not claim to see two noses, but rather my own nose doubled, as John
Searle observed. In the same way, when I see a coin as elliptical, I do
not say I am seeing an elliptical object, but rather a round object that
appears elliptical.[7]
The two
examples just presented highlight the virtues of the critique of language, yet
they also expose its limits. Concerning the first example, it is clear that
even without Plato’s doctrine of Ideas, the problem persists: it is a matter of
explaining how we can say the same of many, that is, the very possibility of
predication, the role of general terms within the sphere of knowledge. One
attempt to resolve this problem is to analyze the meanings conveyed by general
terms as rules of application (Verwendungsregeln), as Ernst
Tugendhat once suggested.[8]
Objections to direct realism
raised through the argument from illusion cannot be dismissed as mere
linguistic confusion. The case for admitting sense-data as the most immediate
vehicles of experience – mediating our access to the external world – has genuine
force. Evidence from fMRI reconstructions of mental imagery suggests that our
immediate visual access is to sense-data.[9]
Yet sense-data are not
themselves the objects we perceive. The mind interprets or “projects” them
outward as belonging to the external world, provided they meet criteria of
external reality such as vivid intensity, independence from the will,
intersubjective confirmation, and conformity to natural laws.[10]
On this account, to say that we
perceive objects directly is to claim that sense-data, situated within the
appropriate context, are apprehended as if they were the external objects
themselves. Their role is analogous to the virtual image of the Moon formed by
a telescope’s eyepiece: although the eyepiece projects a virtual image onto the
retina, what we experience is not the image but the Moon itself. In the same
way, sense-data serve as transparent vehicles of perception, enabling vision
without themselves being the objects of perception.
This view, which I call non-naive direct
realism, preserves the intuitive appeal of direct realism while acknowledging
the indispensable role of sense-data, moving beyond a purely linguistic critique.
2. THERAPY PRESUPPOSES THEORY
Usually, the critique of language is not conceived as a theory of the
nature of philosophy, but merely as a way of doing philosophy. Yet in certain
passages of Wittgenstein’s writings, this critique seems to have become a conception
of philosophy’s very nature. According to some, he envisioned his critique of
language as a form of linguistic therapy, stripped of any positive
content of its own.[11] Even
though I believe Wittgenstein in the end endorsed this view only methodologically
– since he also made observations that distance him from it – such a conception
can be (and indeed has been) drawn from his texts. I shall therefore present it
here solely for what it is able to teach us.[12]
The so‑called therapeutic
conception of philosophy holds that everything carried out under the name of philosophy
arises from linguistic confusions. Philosophers, driven by an irresistible “craving
of generality”, are prone to being misled by the superficial structures of
language, erecting ambitious theoretical “castles of cards.” And when these
constructions collapse into contradictions, those same philosophers find
themselves reduced to hopeless prisoners of “knots of thought.”
Thus, philosophy at its best
should take on a therapeutic vocation: its purpose is to dismantle the fragile
card‑castles of speculative metaphysics and to loosen the knots of thought in
which the most ambitious minds have become entangled. This labor is not
accomplished through the invention of new theories or the explanation of
phenomena, but through attentive descriptions of the ways in which we in fact
employ our words, as it were, “bringing them back from their metaphysical
holidays to their everyday use.”[13] From
this vantage point, philosophy reveals itself as a purely destructive
undertaking, achieving success only when the philosopher, released from
metaphysical anxieties, like the psychoanalytic patient freed from neurotic
fixations, becomes able to forget philosophy itself.
Although one can find many examples
of effective linguistic therapies, it is important to recognize that their
efficacy depends upon theoretical generalizations, even if they remain
implicit. To illustrate, consider the thesis of perdurantism in analytic
metaphysics. From this perspective, material objects are four‑dimensional
entities. They are defined not merely by their three spatial dimensions, but
also by a temporal dimension, namely, their endurance through time. This
conception seeks to explain how a material object—such as an apple or a person—can
remain the same, even as it undergoes significant changes over the course of
its existence.
The practitioner of philosophy
as therapy, in turn, tends to align himself with the intuitions of common sense
and natural language. This leads him to suggest that the perdurantist merely
elaborates upon a primordial confusion between material objects and events
or processes. These last entities truly possess a temporal dimension
that defines them. To illustrate, the process of transformation of a chrysalis
into a butterfly only makes sense when considered as four‑dimensional,
including successive stages. The stance of the philosopher‑therapist, like that
of the critic of language, is here traditionalist and nearly Aristotelian. He
will claim that material objects are three‑dimensional and remain essentially
the same through time, such that only their accidental properties
undergo change, not those enabling us to identify the object as the same
through time. Thus, contrary to the philosopher‑therapist’s expectation, we have
a more intuitive alternative known as endurantism, which also requires theoretical
development, differing only in that it does not rest upon a supposed linguistic
confusion, as the perdurantist’s does.
The difficulty with a strictly therapeutic
conception of philosophy is that it cuts the branches too short. No critique of
language has ever managed to be entirely non‑theoretical or non‑explanatory.
Wittgenstein’s own work is a revealing example of this impossibility, though
the fact is often obscured by the fragmentary and elusive character of his
writings.[14]
This is made clear by the following
example I take from his Philosophical Investigations.[15]
It is framed as a critique of the (reductionist) “label theory” of proper
names, according to which the meaning of a proper name is the object it
designates, and the relation between name and meaning is analogous to that of a
label affixed to a bottle.
However, in rejecting that
theory, Wittgenstein ended up unintentionally sketching his own version of the bundle
theory of proper names. According to him, the meaning of names such as
‘Moses’ is determined by the various definite descriptions attached to them:
“the man who led the Israelites through the desert,” “the man who lived in that
time and place and was called ‘Moses’,” or “the man who, as a child, was drawn
from the Nile by Pharaoh’s daughter.” Using Wittgenstein’s own vocabulary, we might
add that these descriptions are expressions of rules that are helpful in
guiding us in identifying the named object – rules which, taken together, in
some way constitute what we mean by the proper name, its referential sense.
The consequence of this
procedure is that Wittgenstein’s reflections on proper names inevitably take on
a theoretical character, since their therapeutic efficacy depends upon an
implicit generalization of his descriptivism to all proper names. Moreover, these
reflections are unmistakably explanatory, for they covertly aim to illuminate
the mechanisms by which we identify persons through the use of proper names.
This theoretical orientation becomes evident
in the fact that such ideas were later taken up by J. R. Searle[16],
in his formulation of a bundle theory of proper names – an explicitly
theoretical and explanatory proposal – and, in a more decisive way, by the
author of the present book.[17]
Examples such as these show
that a philosophical therapy, if it is to be truly effective – curing the malady
rather than merely easing occasional symptoms – must rest upon generalizations
endowed with explanatory power, even if only implicit. Once developed, such
generalizations compel us to leave behind the terrain of reflections on the
misuse of ordinary language, from which they first arose, and to advance toward
increasingly elaborate theoretical constructions.
Critique and theory, we may conclude, cannot be
wholly separated, for they are like the two opposing sides of the same
philosophical coin. The preference for emphasizing one aspect over the other
tends to be less a matter of principle than of temperament, reflecting each
philosopher’s style and personal inclination.
3. PHILOSOPHICAL ANALYSIS IN ITS BEST
The failure of a purely therapeutic or critical conception of philosophy
compels us to consider the analysis of language. This constitutes the
constructive and theoretical dimension of the philosophical enterprise, capable
of providing a foundation for the critique of language and, indeed, of
transforming that critique into an application of itself. As I have previously
observed, linguistic analysis may be pursued in two distinct orientations: a
formally oriented approach (from the so‑called ‘philosophy of ideal language’)
and a communicationally oriented approach (from the ‘philosophy of ordinary
language’).
An example of the formally
oriented approach was the outline of a general structure of formal language,
evidenced in the distinction introduced by Carnap between formation rules,
responsible for specifying the symbols and well-formed sentences, and transformation
rules, responsible for determining the possible logical relations between
the sentences.[18]
On the other hand, an instance of linguistic analysis conducted under a
communicational orientation is J. L. Austin’s theory of speech acts. In Searle’s
elaboration of this theory, the structure of communicative action is generally
reducible to the schema F(p), where p denotes the propositional
content and F the illocutionary force – the latter specifying the type
of interpersonal commitment that the speaker seeks to establish in relation to
the content (p).[19] For example, in uttering ‘I request that you
close the door,’ the speaker’s act is characterized by an illocutionary force,
namely that of a request, and a propositional content, namely the action of the
addressee closing the door.[20]
Later, Searle developed a
remarkable theory of intentionality in the philosophy of mind, based on the
internalized form of this same schema: what we have is the intention ‘I’
related to an intentional content ‘p’, producing I(p).[21] For
example: if “I” is the intention to close the door, this wanting must be
associated with its intentional content (that the door be closed).
Analytical constructions of
this sort constitute general and wide-ranging theoretical frameworks,
developments that can bring us closer to the horizons of scientific inquiry.
Carnap’s distinction between rules of formation and rules of transformation has
long been assimilated into diverse domains of symbolic logic, which has matured
into a formal science. Similarly, the theory of speech acts now properly
belongs to the field of linguistic pragmatics rather than to philosophy. While
such theoretical architectures may indeed serve as instruments of critique,
this was never their principal aim. Their fundamental purpose has been to
extend the boundaries of human knowledge.
In what follows, I will present
a robust version of the conception of philosophy as the analysis of
language, which will serve as a methodological presupposition for the
subsequent chapters of this book. This is a communication-oriented approach
that stretches to the very limits of tolerance and theoretical defensibility. A
comparable perspective – though with individual variations – can be found in the
work of later and more refined practitioners of analytic methods, such as P. F.
Strawson and Ernst Tugendhat.
A fundamental presupposition of
the more robust conception of philosophy as the analysis of language – already
noted in the preceding chapter – is the idea that we remain largely unaware of
the exceptionally intricate structure of the most central concepts of our
language, and even less of the internal relations that may obtain among them.
Concepts such as truth, belief, perception, knowledge, property, existenz,
number, causality, time, goodness, justice, beauty, and many others, are
intrinsically interconnected, forming a dense conceptual web that we employ in
our everyday discourse.
Wittgenstein, in his Tractatus
Logico-Philosophicus, emphasised this perspective by stating:
Man possesses the capacity of constructing languages, in which every
sense can be expressed, without having an idea how and what each word means
just as one speaks without knowing how the single sounds are produced. Colloquial
language is a part of the human organism and is no less complicated than it.
From it, it is humanly impossible to gather the logic of language immediately.
Language disguises the thought; so that from the external form of the clothes
one cannot infer the form of the thought they clothe, because the external form
of the clothes is constructed with quite another object than to let the form of
the body be recognized. The silent adjustments to understand colloquial
language are enormously complicated.[22]
As has already been observed im the last chapter, this lack of awareness
regarding our conceptual structures has an explanation: we do not acquire these
concepts, nor their possible interrelations, through explicit definitions, but
rather, since childhood, through a non-cognitive praxis grounded in positive
and negative exemplifications, to which our learning is repeatedly subjected by
interpersonal correction. Thus, although it seems evident that we know the meanings
of words such as knowledge, truth, justice, and time, since we are able to use
them correctly, we remain unable to articulate the constitutive rules of their
meanings, that is, the precise contours of their concepts. This is why, despite
our proficiency in their use, we so often find ourselves paralyzed when asked
to explain what we mean by them.
The lack of awareness of the rules governing the use of words central to
our understanding of the world fosters the emergence of philosophical confusion.
Philosophers – especially those devoted to speculative metaphysics – have systematically
misapplied our expressions. Yet it is possible, as Wittgenstein observed, that
such confusions are profound, pointing toward what would otherwise remain
unnoticed.
From the perspective outlined
above, the philosophy of conceptual analysis takes shape as a critical
enterprise, for the very construction of theories that lay bare our most
central conceptual structures already contributes to the dissolution of
conceptual confusions. By involving generalizations, these theories also possess
explanatory value. Their most distinctive purpose, however, should be to
provide what, following Wittgenstein, we might call a panoramic (perspicuous
or synoptic) representation (übersichtliche Darstellung) of
language: a kind of organic theorization that enables us to perceive the
structural relations among concepts of philosophical interest. It is worth
translating the famous central passage:
A main source of our misunderstanding is the
lack of transparency in our grammar. Panoramic representation allows
understanding to perceive the connections. Hence the importance of finding and
inventing intermediate links. The concept of panoramic representation is, for
us, of fundamental importance. It reveals the form of our representation, the
way we see things. (Is this a worldview?)[23]
Since our philosophically significant concepts are, to some extent, interrelated,
panoramic representation may serve to make explicit the systematic relations
among them. Its aim is to elucidate what Tugendhat has termed the conceptual
network (begriffliches Netzwerk), which constitutes our understanding as
a whole.[24]
– The kind of theory on the nature of philosophy that will be developed in the present
book would be classified by Wittgenstein as a kind of panoramic representation
of the grammar of the word ‘philosophy’ (an ‘affresco’ according to an Italian
listener).
To complete
our framework, we must highlight a particularly illuminating feature of
analytic philosophy, identified by W. V. O. Quine as semantic ascent.[25] This
can be understood as a discursive emphasis on the linguistic-conceptual
dimensions of our expressions. Through semantic ascent, these linguistic-conceptual
aspects are brought to the forefront, thereby making subtle distinctions
explicit and helping to prevent confusion.
To illustrate, the question “What are numbers?”
was paraphrased by Frege as “What is the meaning of sentences containing number-words?”,
and Wittgenstein’s assertion “The world is made of facts, not of things” was
reformulated by Carnap as “The conceptual word ‘world’ is understood in such a
way that through it only the system of facts, not that of things, can be
referred to.” The notion of semantic ascent arose, in fact, as a reaction to
Carnap’s related concept of the formal mode of speech, which he regarded
as the proper register for addressing the linguistic-conceptual issues
distinctive of philosophy. However, as Quine perceptively observed, Carnap’s
distinction proves mistaken insofar as he sought to make it a distinctive characteristic
of philosophy. As Quine rightly noted, the notion of semantic ascent differs
from the formal mode of speech in being conceived as applicable not only to
philosophical sentences but to any conceivable sentence. As he wrote:
Semantic ascent applies everywhere. “There are marsupials in Tasmania”
can be paraphrased as “‘Marsupial’ is true of some creatures in Tasmania,” if
there is any point to be made in that. It just happens to be that semantic
ascent is most useful in philosophical contexts.[26]
The notion of semantic ascent, with its linguistic-conceptual emphasis,
can be explained in greater detail by considering that, to avoid errors,
analytic philosophy often presents its arguments more explicitly using a
metalanguage that allows us to focus on the words and concepts themselves.
One way to highlight the whole
function of semantic ascent is to note that it is generally carried out through
a semantic metalanguage, rather than merely a syntactic metalanguage.
This observation allows us to respond to the objection that, as a
linguistic-conceptual enterprise, analytic philosophy would inevitably leave
the world itself aside.
To clarify these concepts,
compare the following two sentences:
(a) ‘Krakow’ is an eight-letter
noun.
(b) ‘Krakow’ is the name of a city located 50° north of the equator and
20° east of the Greenwich meridian.
In sentence (a), we use syntactic metalanguage to treat the word ‘Krakow’
as a physical sign. In sentence (b), we use semantic metalanguage to talk not
only about the word ‘Krakow’, but also about what it means.
Within a Fregean framework, one
may observe that the adoption of a semantic metalanguage serves to make
explicit the senses (Sinne) of our expressions—that is, the modes of
presentation through which their referents are disclosed. Yet in articulating
these senses, we are not confined to the level of linguistic mediation alone;
we thereby also speak of the referentes of these senses, namely, of the world,
insofar as reference—whether to objects or properties—can be cognitively
apprehended only through conceptual articulation.[27]
Such an approach transcends any
reduction to the mere literal meaning of words. Consider, for instance, a
scientific term like ‘entropy’: to speak of its Fregean sense is to
situate the term within its theoretical nexus, thereby elucidating the relation
between the linguistic sign and the manifold ways in which it mediates access
to the world. In this way, semantic ascent does not remove reality from
philosophical discourse but rather clarifies the conceptual conditions under
which reality becomes available to thought.
In summary: Through a syntactic metalanguage, we speak only of signs in
abstraction from their meanings – a procedure that may lead to a barren
formalism. By contrast, a semantic metalanguage preserves meanings together
with their referents, moving beyond the signs themselves. This is the properly
philosophical path, whereby the analysis of language extends from words to what
is meant by them, and thus to the world itself. In this way, it becomes evident
that the conceptual emphasis fostered by semantic ascent is merely a means of
directing our attention to language, without excluding any of the value it may
represent.
Although the formally oriented
mode of analysis – practiced to a considerable extent by philosophers such as
Rudolf Carnap, W. V. O. Quine, Donald Davidson, David Lewis, Saul Kripke,
Hilary Putnam, and David Kaplan – also makes use of linguistic-conceptual ascent,
it nevertheless differs in important respects from the more robust conception
of analysis, particularly in its stance toward the communicative demands
inherent in common sense and in natural language.
Formally inspired philosophers
tend to privilege the internal consistency of their theories, even when this entails
relinquishing consonance with common sense, with the evidence furnished by
ordinary language, or even with empirical science. For this reason, they are
consistently more willing to sacrifice the latter in favor of the former,
exchanging a heuristic burden for an imaginative bonus. David Lewis’s
suggestion of the existence of infinitely many possible worlds—each as real as
incommensurable—is a paradigmatic example of this tendency.
Indeed, many of the ideas arising from the
syntactically inspired approach to the analysis of language stand in manifest
contradiction with certain fundamental intuitions. What accounts for this? I believe
the answer is not difficult to discern. To address it, one must appeal to the
concept of semiotics, which I borrow from Charles Morris.[28]
Semiotics comprises three dimensions: the syntactic dimension,
concerning the relations among signs; the semantic dimension, concerning
the relations between signs and their referents; and the pragmatic dimension,
concerning the relations between signs and their interpreters – a scope that
clearly extends to their effects within communicative interaction. What I intend
to show is that, in general (though the reverse is to some extent possible),
there exists among these dimensions a relation of presupposition, in
which
The
SYNTACTIC
dimension of semiotics
is presupposed by its
SEMANTIC
dimension,
which is in turn presupposed by its
PRAGMATIC
dimension.
We can readily grasp the syntax of a language – that is, the rules
governing the combination of its signs – without any knowledge of what those
signs refer to, without knowing how they are applied in concrete situations,
and thus without even understanding their meanings. The reverse, however, is
far less conceivable: one cannot properly apprehend the meanings produced by
combinations of signs, nor comprehend their referential use in appropriate
contexts, without first knowing their syntactic functions, without knowing how
they may be combined into well-formed sentences. The exception lies in cases
where no syntax exists at all: if a red flag signifies danger, I may understand
its semantic and even its pragmatic dimension without any syntactic dimension,
since none is present.
Summarizing: knowledge of the
pragmatic dimension – understood as the relations of interpreters to signs
within communicative interaction – tends to require prior knowledge of the
constitution of a semantic dimension, namely, the relations between signs and
their referents, since its knowledge is indispensable for communication. Knowledge
of the semantic dimension, in turn, tends to require prior knowledge of the
constitution of a syntactic dimension, that is, a syntactic basis upon which
reference can be effected, together with full communication. Thus, the
pragmatic dimension presupposes the semantic, which in turn presupposes the
syntactic, such that the pragmatic dimension ultimately presupposes both.
This means that the pragmatic
dimension carries, at least implicitly, the assumption of the entire set of
semantic and syntactic rules of language. The pragmatics presupposes the whole
framework that articulates our intuitions about the meanings of our
expressions, as manifested in the ways we use them, alongside the rules of
syntax. Furthermore, the pragmatic dimension is rooted in our form of life, thereby
extending philosophical inquiry in unexpected ways.
This hierarchy carries profound
implications for what I have termed theoretical robustness in analytic
philosophy and helps illuminate Searle’s claim that a communication-oriented
philosophy (such as his own) is “stronger.”[29] Let us now consider the argument.
It is, therefore, an asymmetry with philosophical consequences. It
suggests that a formally oriented analysis, one that emphasizes syntax, is less
dependent on the other dimensions of language. Its treatment of the semantic
and pragmatic dimensions is cursory, if not simplistic, inattentive, and often
accompanied by a reductionist approach. The
result is that such analysis can be pursued without close regard for the other
dimensions and, consequently, may diverge from them without any loss of intelligibility.
Hence, the formally (syntactically) inspired conceptual analyst feels freer to
challenge the assumptions that underlie the rationality of language and our
common understanding of the world, particularly when the very procedure relies
upon a deliberate rejection of those assumptions. In this space, there emerges
a far greater latitude for the procedural stance described in the previous
chapter as reductionism: the tendency to generate ideas and theories
that exclude more than would be intuitively reasonable.
These considerations are of metaphilosophical interest, for they help
explain why many challenging arguments—formulated by philosophers such as
Quine, Kripke, Hilary Putnam, David Kaplan, and David Lewis—were able to oppose
with relative ease the linguistic common sense embodied in ordinary language.
By contrast, philosophers with a communicational orientation, such as John
Searle, P. F. Strawson, and Jürgen Habermas, precisely because they advanced
more robust theories, could do so only at the cost of more readily discernible
inconsistencies. For this reason, their theories remain more vulnerable to the
temptation I have called ampliationism. Both reductionism and ampliationism are
philosophical vices, though there may be advantages in vice, at least so long
as it does not harm one’s intellectual health. To put it more plainly: a
community of ideas, such as our present analytic one, trained for many years in
logical syntax and accustomed to giving only cursory attention to semantics and
pragmatics, will tend to treat the latter in arbitrary ways, without sensing
that it is transgressing the limits of meaningful language.
It should further be observed
that, in its pre‑analytic history – which embraces not only Hegel but Aristotle
as well – philosophy would be more fittingly classified as lying closer to a
communicatively oriented inquiry, with its pragmatic emphasis, than to a
formally oriented inquiry, with its syntactic emphasis. Apart from Leibniz’s
dreams, the formally oriented investigation begins with Frege, the “inventor of
analytic philosophy” (Searle), carried forward by thinkers such as Bertrand
Russell, and was driven to its most radical expression in the formalism of
theorists like David Lewis or, say, Scott Soames, to cite two extreme cases.
Frege himself, however, the founder of analytic philosophy, remarked that his
logic was like a microscope: of great utility in examining microorganisms, but
altogether useless for other purposes. The problem we face today is that of
turning the formalist microscope upon the intricate bloom of an orchid.
In the following section, some
theoretical consequences of the previously outlined conceptions will be
critically assessed. My chief aim is to demonstrate that the conception of
philosophy as the analysis of language – and, by extension, as the critique of
language – while effective in indicating how philosophy ought to be, is
incapable of showing us what philosophy truly is.
4. THE OBJECTAL FALLACY IN ANALYTIC PHILOSOPHY
Since the analytic philosopher is committed to uncovering the conceptual
structure of our language, both the defenders and the critics of philosophy as
conceptual analysis may well think that he:
(a)
Cannot be
engaged in developing speculative hypotheses about the world, as the
traditional metaphysically oriented philosopher once did.
Moreover, it is often thought that he:
(b) Cannot be engaged in the formulation of empirical hypotheses
about the world, as natural scientists do (even if the effort to describe the actual
functioning of language may, in a certain sense, be regarded as an empirical
undertaking).[30]
My aim in this section is to show that neither assertion (a) nor assertion
(b) is borne out by the actual practice of philosophy understood as the
analysis of language, and that such a claim rests upon a subtle objectal
fallacy.
By exposing the fallacious character of these
assertions, I also intend to demonstrate that the assumption is mistaken that,
from the standpoint of its object of inquiry, analytic philosophy would be
essentially distinct from other forms of investigative activity. After all,
although its purpose is the clarification of conceptual structures, we have already
seen that this does not entail renouncing any explanatory claims about the world.
The convoluted argument I will
present to support this point is, I admit, not a model of linearity or
transparency. Anyway, here it goes:
To demonstrate that the
conceptual analyst fails in ensuring that his analysis has a distinct object of
investigation, different from that of traditional philosophy or even science in
general, we must begin by considering his actual practice. Theses (a) and (b)
could, indeed, be consistently maintained if the conceptual analyst confined
himself to the logical analysis of sentence structures or to a tedious,
quasi-lexicographic description of the uses or meanings of philosophically
relevant conceptual words in our natural language, as was the case with much of
the so‑called ordinary language philosophy.
Yet this is not what truly
happened. In order to attain philosophical relevance, the conceptual analyst
was very often compelled to go further: it was necessary to inquire into our
real praxis of thinking about things, even to the point of identifying within
that praxis concepts for which no words yet existed in the language. Such
concepts are selected according to criteria such as the enhancement of coherence
and the explanatory power of their theoretical sketches.
To show that the conceptual
analyst has not succeeded in securing for his analysis an object of
investigation distinct from that of traditional philosophy or even of a
possible scientific endeavour, we must begin by considering his actual
practice. Theses (a) and (b) could, in fact, be consistently upheld if the
conceptual analyst confined himself to the logical analysis of sentence
structure or to a tedious, quasi‑lexicographic account of the uses or
meanings of philosophically relevant conceptual words in our natural language,
as was the case with much of the so‑called ordinary language philosophy.
Yet this is not what truly
occurred. In order to attain philosophical relevance, the conceptual analyst was
very often compelled to go further: it became necessary to inquire into our
real praxis of thinking about things, even to the point of identifying
within that praxis concepts for which no words yet existed in the
language. Such concepts are chosen according to criteria such as the
enhancement of coherence and the explanatory power of their theoretical
sketches.
Since these newly discovered concepts can only be expressed through
novel combinations of words, the conceptual analyst is often led to replace such
combinations with terms of art, devised for reasons of discursive
economy. A few examples illustrate this procedure: (i) the proponent of a
theory of communicative action may analyze our speech acts in light of
their illocutionary force; someone engaged in the philosophy of content
may seek to understand the representational function of declarative sentences,
their cognitive meaning in terms of verification rules; a
theorist of reference may distinguish the proper name from the definite
descriptions associated with it by calling the former a rigid designator;
and an epistemologist may propose an analysis of the concept of propositional
knowledge (knowing that) as “Knowledge is not simply
justified true belief, but it is justified true belief, justifiably arrived at”[31] (see
Appendix I)
In reflecting on the matter, a
first point to consider is that the supposedly analytic procedure may, in fact,
involve moments of synthesis. To illustrate: when we perceive that concepts C1
and C2 are internally related, or that both are constituents in the analysis of
C3, we are already engaged in an act of synthesis. Moreover, deep conceptual
structures may first be discovered and only thereafter analyzed and
named. In this process, the philosopher is already engaged in a work of generalization.
In other words, he may seek to bring to light the intermediaries that, at least
for the pragmatically oriented analyst, must already be implicitly given in the
applications of our language.
A difficulty is that the adequacy
of these supposedly discovered unifying concepts may be highly hypothetical.
This becomes evident in the recurring controversy surrounding the meanings of
the general terms employed to account for a new conceptual unity. Often, the
philosopher is engaged in establishing novel concepts and relations, seeking to
justify them by their coherence with the broader system of beliefs he accepts.
Such an undertaking is inevitably conjectural. Yet in this pursuit of reflective
equilibrium, the philosopher can advance hypotheses that may prove genuinely
fruitful.
Let us consider some examples of such conceptual hypotheses formulated
within the domain of analytic philosophy. They concern: (i) the empirical
structure of language, as in the theory of speech acts; (ii) the representational
function of our statements, as in the more speculative case of the principle of
verifiability, according to which the cognitive meaning of a declarative
sentence consists, following Wittgenstein’s exposition, in multiple modes of
verification of greater or lesser strength[32];
(iii) the manner in which the mind assesses our “knowing that p” in the
definition of propositional knowledge.
All these cases may be regarded, to some
extent, as analogous to the work of discovering natural laws in the empirical
sciences – that is, analogous to something capable of explaining an
indeterminate multiplicity of particular cases, later to be confirmed by
experience, whether this concerns the rules of linguistic interaction in the
first case, or the structure of essentially cognitive processes in the latter
two.
I believe that a philosopher of
communicational orientation will have little difficulty in accepting these
conclusions. Nevertheless, some will tend to insist that, although his concrete
analytical procedure is preceded by a hypothetical moment of synthesis, his
effort is always directed toward making explicit what already belongs to our
conceptual system, and not, as in the case of the empirical scientist or the
speculative philosopher, toward surpassing it through the elaboration of
hypotheses about the actual empirical world.
Yet, when we examine the
available examples, we perceive that much of what analytic philosophers say can
also be interpreted as dealing with empirical facts, even if these are highly
general, or concern above all the manner in which our representations relate to
the world, rather than the world itself.
Indeed, when we examine
examples of analysis that go beyond the philosophy of language – such as those found
in the philosophy of mind or in analytic metaphysics – we realize that they may
well refer to the general constituents of the empirical world itself. Entities
such as properties, individuals, causes and effects, space and time, not to
speak of action and, maybe, justice and goodness, are highly general
constituents of the external and human world, and the investigation of their
nature is not devoid of empirical character.
To illustrate, let us consider
the analysis of the concept of consciousness in the philosophy of mind. I shall
restrict myself here to one of the earliest among the many theories of
consciousness developed over the past five decades: the theory proposed by D.
M. Armstrong in 1978, a philosopher trained in the analytic tradition.[33]
Armstrong introduced a fundamental distinction between two principal forms of
consciousness: perceptual consciousness, which corresponds to
wakefulness and the perception of the world, and introspective consciousness,
which consists in the submission of mental states to introspection or to second‑order
cognition, thereby making them conscious. Although this distinction may be conceptual,
it also refers to classes of empirical phenomena, namely, mental phenomena that
are diffusely situated in space and time.
Within the domain of analytic
metaphysics, the ontological theory of tropes – first articulated in 1953 by
the metaphysician Donald Williams – has assumed a central place in contemporary
debate. This proposal reimagines the very notion of property, departing
from the venerable realist tradition that conceives of properties as abstract
universals. Williams, instead, advanced a striking alternative: the world, he
argued, is constituted not from transcendent abstractions but from spatio-temporally
localizable properties, which he named tropes.[34]
I am not alone in defending the
view that the conceptual analyst may lay bare empirical structures. Many years
ago, A. J. Ayer likewise aligned himself with this perspective:
The distinction between “about language” and “about the world” is by no
means abrupt, for the world is the world we describe – the world as it figures
within our conceptual scheme. In exploring our conceptual scheme, you are at
the same time exploring the world itself.[35]
Ayer’s response becomes particularly compelling in light of our earlier
reflections on conceptual emphasis and the employment of a semantic
metalanguage. It gestures – albeit implicitly – toward the further point that
the proper object of philosophical inquiry cannot be distinguished solely on the
basis of analyzing our traditional conceptual structures in the independence of
the development of the sciences. For, in an analogous sense, one might argue
that both the empirical scientist and the speculative metaphysician are engaged
in a kind of “conceptual analysis.” The difference, however, is that the
empirical scientist has no need for devices such as semantic ascent or the
deployment of a semantic metalanguage in the course of his work. This final
point may be clarified still more effectively if we examine separately the
objections raised against theses (a) and (b).
Consider thesis (a): according
to it, unlike speculative philosophers, analytic philosophers do not advance
conjectural assertions about the world. Against this thesis, it is important to
emphasize that the history of philosophy has shown that every domain and
position of the traditional discipline can be found within the work of
philosophers regarded as analytic (by some even described as “post-analytic”). It
scarcely makes sense to maintain that analytic philosophy is not speculative,
for history itself reveals that the distinctions upheld by different
philosophers between two kinds of philosophy:
IMMANENT METAPHYSICS
vs. TRANSCENDENT METAPHYSICS
(restricted to the
world of) (seeking to go
beyond the senses,
the senses) relating to the supra-sensible world)
(W. H. Walsh)
CRITICAL PHILOSOPHY
vs.
SPECULATIVE PHILOSOPHY
(concerned with the
definition (aiming
to reach general conclusions
and critical analysis of the about the nature of the
universe and
concepts of our daily life about our place and expectations
and the sciences) within it)
(C. D. Broad)
DESCRIPTIVE
METAPHYSICS vs. REVISIONARY METAPHYSICS
(concerned with the
description (attempting to
create a new
of our actual structures of structure of thought)
though)
(P. F. Strawson)
These distinctions find a parallel in the field of analytical
philosophy, in the distinction between:
The results of the analysis vs. The results of the analysis of formally
of communicationally oriented language.
oriented philosophy (syntatic and semantically
oriented
(pragmatically oriented philosophy)
Philosophy)
Although this parallel is not rigorous, there is a profound reason for
its existence. The reliance upon common-sense intuitions – often mirrored in
linguistic intuitions – typically safeguarded by critical philosophy and by
immanent, descriptive metaphysics, finds its counterpart in the stance of the
conceptually oriented analyst inspired by communicative practice. By contrast,
the formally oriented conceptual analyst shows little or no concern for
preserving such ordinary intuitions, whether or not they are reflected in
language, thereby enabling a far more speculative undertaking.
Consider, as examples of formally oriented
conceptual analysis, Wittgenstein’s Tractatus Logico-Philosophicus, with
its theory of atomic names referring to absolutely simple, indestructible objects
as their meanings – a reflex of the supposed logical proper names inexistent in
our natural language. Consider also Saul
Kripke’s causal-historical theory of reference, according to which proper names
identify their referents because they are at the end of the appropriate
external causal chain initiated by an act of baptism,[36] –
although we cannot identify any link of it without circularly already knowing the
name’s reference. Consider, finally, David Lewis’s modal realism, which boldly
posited the existence of an infinity of real, mutually incommensurable possible
worlds – appealing to an unverifiable hypothesis.
These observations demonstrate
that the distinction between analytic and traditional philosophy does not apply
to the object of investigation. Indeed, if we are sufficiently
imaginative, all speculative metaphysics may be reformulated appealing to semantic
ascent, employing a metalanguage capable of legitimizing the speculative
philosopher’s claim to conduct philosophical analysis on the very same terms as
any conceptual analyst.
Consider, as a radical example,
the concept of the pure ego in Fichtean idealism. It is an entity accessible
only intellectually, which posits (setzt) the external world in order to
posit itself (by Selbstsetzung), while simultaneously establishing a necessary
opposition to it, from which there follows, as synthesis, the finite external and
internal world. Now, it would hardly be surprising if some contemporary
analytic philosopher, sympathetic to idealism, were to translate such claims
into an analysis of the concept of the “elusive self,” understood both as
constituted by and as constituting social reality under anti-realist
assumptions.
Even if such anti-realism ultimately proves
to be as scarcely intelligible and as speculative as the Fichtean model itself,
it would be no less defensible than some ideas of contemporary social constructivism
in the philosophy of science.[37]
Although this type of strategy can be readily implemented by the
formally inspired conceptual analyst, we have already seen that it would demand
greater effort from the communication-oriented conceptual analyst, since it
tends to conflict with the pragmatic common-sense intuitions often reflected in
ordinary language, and even with those of science, without providing sufficient
justification for such a rupture. The maxim of epistemic integrity here is: “do
not transgress ordinary language unless you have sufficient reason to do so”.
Nevertheless, even in such
cases, the strategy is not unworkable: the communication-oriented analyst may
argue that the conflict with our intuitions is merely apparent, seeking
to demonstrate that his claims can be reconciled with the broader background of
our ordinary beliefs. After all, Bishop Berkeley, in defending the view that
our world is constituted solely by ideas and spirits, anticipated this very
strategy when he asserted that his immaterialism did nothing more than reflect
the genuine expectations of common sense – those preserved by individuals still
untouched by philosophy.
To summarize: the work of
analytic philosophers includes moments of hypothetical synthesis, in
which new concepts are conceived and explored in their possible interrelations.
Their work is capable of containing, even if only indirectly, unexpected
metaphysical speculations, which may have consequences for the very way in
which we ground our apprehension of empirical reality.
By contrast, the formally
oriented conceptual analyst tends to engage in such speculations with greater
ease, for he is more willing to sacrifice alignment with intuitive expectations
about the world without thereby compromising the intelligibility of his
arguments. This is because, for him, intelligibility rests above all on the
internal formal coherence of his analysis.
In light of these considerations, it becomes evident that all domains of
traditional metaphysics can, in one way or another, be approached through an
expanded form of linguistic-conceptual analysis, capable of achieving something
as seemingly trivial as semantic ascent. The conclusion is inevitable: to
maintain that there exists any substantial distinction between philosophy as
conceptual analysis and traditional philosophy—even in its most speculative
strands—is to reify the merely instrumental role of linguistic-conceptual
emphasis, or, to put it differently, of tools like the Fregean logic or the
semantics of possible worlds.
A similar argument applies to
thesis (b), according to which philosophy differs from the empirical sciences
by restricting itself to conceptual investigation.
That this thesis is false
should already be evident, since our last example of conceptual analysis also
involved the natural world, albeit indirectly. Yet the point at issue can be
formulated more dramatically.
Let us first suppose the
existence of a fully consistent conceptual analyst who, by adopting the broad
conception of analysis previously described, regards concepts and their
relations as the true objects of philosophical inquiry—those which distinguish
philosophy from other domains of knowledge. How, then, would such an analyst
view science?
It would not be difficult for
him to perceive that Einstein, in concluding that the speed of light is a constant
for all observers, was compelled to analyze the concept of simultaneity
as applied to observers in relative motion at velocities approaching that of light.
For it is certain that he was not attempting to analyze the relative motion of
people, cars, and trains in our everyday surroundings.
With regard to the work of the
cosmologist Stephen Hawking, our conceptual analyst would readily observe that
the scientist was not engaged in any literal dissection of black holes themselves,
but rather in rigorous astrophysical analyses concerning the implications of
the very concept of the black hole – implications indispensable for a
coherent interpretation of the phenomenon.
In the same way, the concept of
natural evolution – as our perfectly consistent conceptual analyst would
soon recognize – was first formulated and properly analyzed by Charles Darwin,
through reflections grounded in zoological and botanical observations. Gregor
Mendel analyzed the concept of the gene; Watson and Crick analyzed the concept
of DNA, abstracting from the process of reproduction. The psychologist Carl
Jung envisioned and analyzed the concept of the collective unconscious:
its archetypes were already present in his patients—it was merely a matter of
naming them. And the sociologist Thorstein Veblen analyzed the concepts he himself
made explicit: that of the leisure class (which prides itself on not
needing to work) and that of conspicuous consumption (the habit of
elites to consume costly and superfluous goods and services as a form of
display).
Were all these persons engaged in
doing philosophy? Granting, as our thoroughly consistent conceptual analyst
does, that the conceptual world is the true object of philosophy, he cannot
avoid an affirmative answer. Indeed, within that horizon, every theoretical
enterprise seems, in one way or another, to become a matter of conceptual
analysis, though not necessarily philosophical in nature.
However, we can imagine the
opposite situation: suppose we have before us a radical empiricist who begins with
the premise that empirical scientific knowledge is not essentially conceptual,
even though he accepts that it can be accessed only through concepts, since
those concepts apply to empirical facts, however broad they may be. How would
such an empiricist regard most of the questions raised by philosophy?
Consider our examples. The
theory of speech acts deals with human communicative actions in real contexts;
the verificationist analysis of the factual or cognitive sense of our
statements concerns the ways our minds establish truths about the world; and
realism about scientific laws is a thesis about the ultimate constitution of
reality. In light of this, the analyst would be led to conceive much of
philosophy as concerned with empirical phenomena – diffuse and wide-ranging, to
be sure – but nonetheless phenomena to be investigated by empirical science.
The case of the fully consistent
conceptual analyst shows that an inquiry which does not directly concern
concepts—such as empirical science—can always be interpreted as involving
conceptual content. On the other hand, the case of the radical empiricist
reveals that an inquiry usually conceived as centered on concepts – such as
that practiced by so‑called analytic philosophers – can, in many instances, be
understood as a form of investigation that transcends conceptual boundaries and
enters the domain of empirical knowledge.
What conclusions can we draw
from this? The first is that the objects of analytic philosophy need not
differ, by their very nature, from those of traditional speculative philosophy,
nor from those of science. After all, analytic philosophy cannot claim an
essential distinction from these enterprises merely by emphasizing the work
with conceptual structures.
Consequently, our two cases
demonstrate, quite unequivocally, that any attempt to restrict the object of
philosophical inquiry to the examination of the structure of our concepts, when
properly construed, ends up undermining an objective distinction between
analytic philosophy and other theoretical enterprises.
1. CONCLUSION: A PARALLEL WITH ARISTOTLE’S ORGANON
So what, then, is the real difference between philosophy as conceptual
analysis and, say, traditional speculative philosophy or even empirical
science? It cannot just be the object of investigation, because as we’ve seen,
the objects overlap all the time.
The difference is
methodological. Analytic philosophers impose a much stricter discipline on their
inquiries. They don’t just spin out ideas in ordinary language; they formulate
their conceptions in a semantic metalanguage, and increasingly they test them
with formal tools—logic, linguistics, even computational models. And here’s the
crucial point: they don’t do this in a vacuum. Their conceptions are constantly
measured against the background of our contemporary worldview, a worldview itself
saturated with advances in modern science.[38]
So, what follows from all this?
Well, the point is fairly straightforward. Analytic philosophy – especially
what used to be called “the philosophy of linguistic analysis”[39] – is really
just the name we give to a more refined way of doing philosophy that emerged in
the twentieth century. It’s not a different subject matter; it’s not a
different universe of problems. It’s a different style. And that style puts a
special emphasis on the language and conceptual medium, and it does so for
reasons of methodological rigor.
Now, think of philosophy as a
kind of argumentative game. You don’t play it with chess pieces, you play it
with symbols, with language. And the goal – indirectly, but unmistakably – is truth.
Once you see it that way, it becomes obvious why new procedural instruments,
new formal tools, became so central to philosophy in the last century. You
can’t play the game well unless you sharpen the tools.
This is most evident in the
core theoretical areas, such as epistemology and metaphysics, as well as in the
philosophies of language, science, mathematics, and logic. In fact, there’s a
revealing historical parallel here. Aristotle, in his Organon, gave
philosophy a set of logical and methodological doctrines that he thought were
indispensable for serious reasoning. He gave us a theory of propositions, a
theory of deductive reasoning (syllogistic), reflections on definitions,
rudiments of induction, scientific explanation and prediction, and even a
classification of fallacies.
The assimilation of these
doctrines was slow. But gradually, they decisively transformed the ways of
doing theoretical philosophy. Aristotle’s instruments of investigation were
refined in the universities of the late Middle Ages, especially under the name
of dialectic, which established new and irreversible argumentative standards,
at least in theoretical philosophy – standards that, once adopted, could never
again be entirely ignored, at least in the domains to which they applied.
Now, what we call analytic
philosophy can be explained as the consequence of a similar methodological revolution.
Since the end of the nineteenth century, there have been enormously important
developments in areas very much like those covered by Aristotle’s Organon.
Some concerned the structure of propositions (as in Fregean semantics), others
deductive logic (first- and second‑order predicate logic, and later modal logic),
others inductive reasoning (probability theory, descriptive decision theory),
still others concerned pragmatics (verification theory, speech act theory), and
finally the philosophy of science (theories of explanation, of confirmation).
It would be genuinely
surprising if philosophical language, at least in many of its domains, had not
been profoundly reshaped by these developments. They established higher standards
of clarity and rigor, and they greatly increased philosophy’s heuristic power, so
long as they were properly applied. The incorporation of these new procedures
into scientifically informed inquiry inevitably made possible a sharper vision,
a revolution comparable to what the telescope meant for astronomy. (Though, as
it is good to remember, the telescope is only for the celestial bodies, not for
what lies immediately around us.)
Now, let’s recap the main
results. The deep reason analytic philosophy once seemed to have language as
its sole object lies in its preparatory concern with the linguistic‑conceptual
element. This concern became especially evident in what Quine called “semantic
ascent.” And here’s where the mistake crept in: analytic philosophers initially
confused the new instruments and procedures of approach, which can also be
applied elsewhere, with the very method of philosophy itself. They then made
the further mistake of taking the object of application of those instruments as
the proper object of philosophy.
The fact that, in philosophy,
we often resort to a semantic metalanguage in order to treat
linguistic‑conceptual structures with greater rigor does not mean that we
ignore their senses and references – conceived, after all, through those very structures.
Nor does it compel us to relinquish access to the world. This resource, like
others, is simply part of what has somewhat misleadingly been called “analytic
philosophy of language.” And that label is also misleading because, in addition
to panoramic representations, it also involves an inevitable component of
theoretical synthesis.
Indeed, if “conceptual
analysis” names anything, it names modes of investigation that combine a more
refined semiotic awareness with heuristic procedures that became common in the
central domains of philosophy throughout the twentieth century, beginning with
Gottlob Frege, the “inventor of analytic philosophy” (Searle). More broadly,
one can say that the so‑called analytic philosophy resulted, to a large extent,
from the development of semiotic sciences applied to philosophy. And that has
nothing to do with finding philosophy’s own inalienable method.
In sum, “conceptual analysis”
is the name given to the most salient procedural traits of a historically given
state of the art. It is, if you wish, a style we shall endeavor to preserve
throughout the course of our inquiry.
If with this rather tedious
chapter, we have not advanced in any significant way, we have at least freed
ourselves from some worries.
[1] To appreciate the extraordinary variety of
methodological and heuristic tools employed by contemporary philosophers, one
might turn to The Oxford Handbook of Philosophical Methodology, edited by
Herman Cappelen and others (2016). While useful, the book illustrates a recurring
limitation in today’s philosophy: a deferential reliance on the wisdom inherited
from Anglophone analytic philosophy over the past five or six decades. Major
figures such as Wittgenstein, Russell, Husserl, and even Gottlob Frege are seldom
mentioned. Meanwhile, essential critics—like John Searle, a staunch opponent of
formalism, and Susan Haack, an outspoken critic of the current state of analytic
philosophy—are virtually silenced.
[2] Ernst Tugendhat, Die Philosophie unter
den sprachanalytischen Sicht, p. 268.
[3] W. V. O. Quine, Word and Object, p.
270 ss.
[4] W. V. O. Quine, “A Letter to Mr. Ostermann.”
[5] Ver Timothy Williamson: The Philosophy of Philosophy,
cap. 1.
[6] J. O. Urmson Philosophical Analysis:
Its Development between the Two World Wars.
[7] The classic ordinary language critique of
the argument from illusion is found in J. L. Austin’s book Sense and Sensibilia.
A very sharp, though schematic, critique of the argument from illusion can be
found in J. R. Searle’s book Language, Mind and Society: Philosophy in the
Real World, Chapter I, p. 28 ff. Searle’s final word in defense of direct
realism is in his highly recommended book, Seeing Things as They Are (2015).
[8] Vorlesungen zur Einführung in die sprachanalytische
Philosophie, pp. 484-488. See also Claudio Ferreira-Costa, How Do Proper
Names Really Work? pp. 220-241.
[9] Rakhimberdina, Z., Jodelet,
Q., Liu, X., & Murata, T. “Natural Image Reconstruction From fMRI Using
Deep Learning: A Survey.”
[10] Cf. Claudio Costa, “The Skeptical Deal With our Concept of External Reality”.
[11] Ludwig Wittgenstein, Philosophical Investigations
I, sec. 109, 118, 119... A well‑articulated, though interpretatively
objectionable, attempt to defend this deflationary interpretation of
Wittgenstein’s view can be found in Paul Horwich, Wittgenstein’s Metaphilosophy.
[12] I prefer to think that, in writing about ‘therapy,’
Wittgenstein was speaking of his personal and minimalist way of working with
philosophical concepts, and not that he was proposing the one and only proper
method of philosophizing. Only this explains why he also maintained ideas that
were different and apparently incompatible with it, as his best interpreters
have pointed out. See, for example, Anthony Kenny, “Wittgenstein and the Nature
of Philosophy”. See also Claudio Costa, Wittgensteins Beitrag zu einer
Sprachphilosophischen Semantik.
[13] Wittgenstein, Philosophical Investigations,
part I, sec. 116.
[14] As A. J. Ayer critically observed: “His repeated preference for description
over explanation, and his abstention from theory – which he claimed to practice
and delighted in presenting to his readers – are not features of his actual
procedure at any stage of his development, including that of the Philosophical
Investigations. That his explanations are runic does not reduce them to mere
descriptions: his theories do not cease to be theories simply because they are
covertly laid down.” Ludwig Wittgenstein, p. 137.
[15] L. Wittgenstein,
Investigações Filosóficas, sec. 79.
[16] J. R. Searle, “Proper Names.”
[17] How Do Proper Names
Really Work?
[18] R. Carnap, Logische Syntax der Sprache, parte I.
[19] J. R. Searle, Mind, Language and Society:
Philosophy in the Real World, p. 138.
[20] Both components can be alternately implicit: I can say
“Close the door”, and the context will indicate whether it is a request or an
order. I can say “Please” and the context will show the listener that it is a request
to close the door..
[21] J. R. Searle, Intentionality: An Essay in the Philosophy
of Mind, cap. 1.
[22] Tractatus Logico-Philosophicus 4.002.
[23] Philosophische Untersuchungen I, sec. 122. On the concept of
perspicuous or panoramic representation (übersichtliche Darstellung), see
the remarks of G. P. Baker & P. M. S. Hacker in Wittgenstein: Understanding
and Meaning, p. 489. (I am referring to the first edition of the text, which
I prefer to the revised version prepared solely by Hacker after the death of his
coauthor.)
[24] Ernst. Tugendhat, “Die Philosophie unter den sprachanalytischen
Sicht“, p. 268.
[25] W. V. O. Quine: World and Object, p.
[26] W. V. O. Quine, Word and Object, pp.
271-272.
[27] Frege described the sense of a name as die
Art des Gegebenseins eines Gegenstandes—the “mode of presentation of the
object.” However, he did not undertake an analysis of the sense of predicative
expressions.
[28]See Charles Morris, Foundations of the Theory
of Signs.
[29] Searle remarked in class that his
pragmatic-communicational approach to philosophical questions is “stronger” than
an approach of purely formal inspiration.
[30] Kai Nielsen emphasised the obvious but remarkable fact
that when philosophers describe the uses of our expressions, “they are making
empirical observations about how language works.” “What is Philosophy?”
[31] See Robert Fogelin, Chapter I of his book Pyrrhonian
Reflections on Knowledge and Justification. See also Chapter V of my own
book Lines of Thought: Rethinking Philosophical Assumptions. Appendix I summarizes
Chapter V.
[32] The original suggestion that the meaning of
a declarative sentence consists in its procedures of verification was made by
Wittgenstein, and it has very little to do with the untenable formalist solutions
later attempted by members of the Vienna Circle – a straw man they themselves
created and, quite rightly, later rejected. Wittgenstein’s original suggestion
requires a semantic-pragmatic development which, as far as I know, has never been
seriously undertaken. See Wittgenstein’s Lectures, pp. 28–29 (sec. 24).
For a more detailed discussion, see Claudio Costa, Philosophical Semantics:
Reintegrating Theoretical Philosophy, chap. V., which is summarized here in
Appendix II.
[33] See D. M. Armstrong’a summary of his theory
in: The Mind-Body Problem: An Opinionated Introduction, cap. 10.
[34] Donald
Williams, “The Elements of Being”, parts I e II. The theory has been gradually diluted
and transformed, so it is advisable to return to the original.
[35] Ayer, in his interview with Brian Magee (Men
of Ideas, p. 127). Magee’s objection to this remark by Ayer – and to similar
observations made in his interview with Searle, an objection to which I respond
here in greater detail – was that analytic inquiry, like any metalinguistic inquiry,
inevitably leaves the real world aside (see Brian Magee in Confessions of a
Philosopher, pp. 74–76). Others, such as John McDowell, have also noted
this, though under the admission of an unwarranted semantic externalism (Mind
and World, p. 27). I say unwarranted because meaning belongs primarily
to language no less than truth. We say that the Industrial Revolution had great
meaning for the evolution of humanity, but here the word ‘meaning’ appears
merely as a substitute for importance. My critique of the externalism of meaning,
as developed by Hilary Putnam, is summarized in Chapter VI, sec. 7 and 8 of my
book How do
Proper Names Really Work?
[36] It should be noted that these external
causal chains do indeed exist. The problem is that they do not, as such,
possess any explanatory power. See Claudio Ferreira-Costa, How do Proper Names
Really Work? Chap. II.
[37] One example of this is the book by B. Latour
& S. Woolgar, Laboratory Life: The Construction of Scientific Facts.
They had the idea of conducting field research investigating the daily routine
of a scientific laboratory as if they were studying the rituals of shamans in an
indigenous community.
[38] A historical exception seems to have been
the philosopher G. E. Moore. However, he may be regarded as a late heir of the old
school of common sense, living among the analytic philosophers and often being
mistaken for one of them.
[39] The swan song of communication-inspired philosophy was
Ernst Tugendhat's classic 1976 study, entitled Vorlesungen zur Einführung in
die sprachanalytische Philosophie. Since then, formally oriented philosophy
has increasingly become almost hegemonic.
Nenhum comentário:
Postar um comentário