1. Introduction

The putative understanding of Rudolf Carnap's Der logische Aufbau der Welt Footnote ¹ (hereafter, Aufbau) as a rigorous defense of empiricist epistemology has lost its credibility due to a number of revisionist interpreters who have emphasized the neo-Kantian aspects of Carnap's first major work (e.g., see Coffa Reference Coffa1991, chaps. 11–12; Friedman Reference Friedman1999, chaps. 5–6, Reference Friedman2000, chap. 5; Haack Reference Haack1977; Richardson Reference Richardson1998; Sauer Reference Sauer1989). Recently, Michael Friedman and Alan Richardson have made decisive contributions in this regard, arguing persuasively that, in the Aufbau, Carnap is addressing the Kantian problem of how knowledge is possible and articulating a structuralist notion of scientific objectivity. Friedman and Richardson forcefully reject the so-called ‘received view’ of the Aufbau in which Carnap's empiricist-reductionist project is motivated to specify a privileged basis of certainty in an empirical foundation, a view that they claim has gained prevalence due to Quine's ([1951] Reference Quine1980, Reference Quine and Quine1969) well known criticisms of Carnap's project (see Friedman Reference Friedman1999, 89–90; Richardson Reference Richardson1998, 10–13).

The aim of the present study is to re-examine the nature and motivations of Carnap's construction theory in light of the revisionist outlook on the Aufbau. My main contention is that the method of rational reconstruction proffered in the Aufbau is intimately linked with Carnap's attempt to advance a method for the justification of conceptual knowledge. The paper proceeds as follows. In Sections 2 and 3, I explicate the nature of construction theory with reference to the revisionist interpretation of the Aufbau. Central themes discussed in these sections include reductionism, objectivity, structuralism, and conventionalism. In Sections 4 through 6, I argue that construction theory is Carnap's proffered method of epistemic justification. Since Carnap endorses a restricted sense of ‘justification,’ the main purpose of the latter portion of the paper is to articulate what justification in the Aufbau amounts to, focusing on both its conventionalist and empiricist aspects. It should be stated at the outset that my aim in this paper is not to undermine the revisionist interpretation of the Aufbau, but rather to indicate its limitations and elaborate it.

2. Construction Theory: A Rational Reconstruction of Concepts

The epistemological method proposed by Carnap in the Aufbau, construction theory (Konstitutionstheorie), is demonstrated through the presentation of a particular constructional system (Konstitutionsystem). In the constructional system outlined in the Aufbau, Carnap reduces all concepts to a basis of subjective experience (§§ 106–156), ‘concept’ (or ‘object’) being crudely defined as “anything about which a statement can be made” (§ 1).Footnote ² In the preface to the second edition, Carnap writes:

To achieve the desired clarity and exactness, Carnap employs Whitehead and Russell's (Reference Whitehead and Russell1910–1913) theory of relations to the task of ‘analyzing reality’ (§ 3). In providing an analysis of ‘reality,’ Carnap means to be analyzing a ‘constructional’ or ‘empirical’ concept (as opposed to a metaphysical concept) of reality, conceived of as a practical heuristic for analyzing the empirical sciences (§§ 170–178). The result of such an analysis is an epistemological genealogy of concepts in which each concept has a definite place (§ 1). A scientific concept (e.g., ‘atom’) is reduced when every statement containing the concept is transformed into a statement containing concepts referring to the immediately given, thereby constituting the concept. Such constitutional or constructional definitions form a hierarchy, with undefined subjective experiential concepts at the ‘ground’ (auto-psychological) level and concepts of increasing complexity at ‘higher’ (physical, heteropsychological, cultural) levels. Concepts at higher levels are to be derived stepwise from the fundamental concepts (§ 1). The resulting constructional system is expressed in the language of modern symbolic logic.

In the Aufbau, reductionism is the central methodological strategy utilized by Carnap to transform all concepts into concepts referring to sense-data (§ 2). Carnap states that a concept is reducible to another concept if all statements about the former concept can be transformed into statements about the latter concept; A is reducible to B, if statements about A can be transformed into statements about B (§ 2). The reduction of a concept involves the production of a rule that prescribes how statements about concept A are to be transformed into statements about concept B. The general rule that indicates how a concept is to be transformed into another is called a ‘construction rule’ or ‘constructional definition.’ Carnap defines ‘transformation’ more precisely with the notion of coextensiveness (i.e., universal equivalence) of propositional functions (§§ 32, 35).Footnote ³ A proper transformation (and hence reduction) of concept A to concept B occurs if for each propositional function exclusively about concept A, there exists a coextensive propositional function exclusively about B (§ 35). Conformably, Carnap defines ‘constructional definition’ as a rule of translation that suggests how any propositional function in A can be transformed into a coextensive propositional function in which only B occurs. The simplest type of constructional definition employed in the Aufbau is the rule of explicit definition, which prescribes that all occurrences of A in a statement are to be replaced with equivalent statements involving B.

On the received view (i.e., the view suggested by Quine), the purpose of Carnap's reductionist project is to ground scientific concepts in a (phenomenal) basis of certainty. Revisionists rightfully eschew this view, emphasizing that Carnap's utilization of a phenomenal basis in the Aufbau is circumstantial rather than essential to construction theory. Carnap merely favors this basis for its ‘epistemic primacy’ and alternative bases (e.g., physical or heteropsychological) can serve as equally legitimate (§§ 57–63).Footnote ⁴ Carnap's qualification that it is possible to reconstruct concepts using alternative bases anticipates the ‘liberalization of empiricism’ associated with his ‘principle of tolerance’ (Carnap [1934] Reference Carnap and Smeaton1937, § 17). In his autobiography, Carnap (Reference Carnap1963a) is especially clear on this issue:

Besides this tolerant attitude towards possible bases for constructional systems, it is significant that Carnap makes no indication that the auto-psychological basis that he uses in his trial system is more certain than any of the other domains that are constructed from it (§ 106). On the contrary, all of these levels are informed by the empirical sciences and equally subject to revision (§ 122). Based on these considerations, revisionists argue that the aim of the Aufbau is not to rationally reconstruct scientific concepts using a phenomenalist basis, per se, but more generally to rationally reconstruct concepts using any basis. On this understanding, it is misguided to view Carnap's project as being motivated to identify an empirical foundation of certainty for epistemology in the tradition of British empiricism. On the contrary, construction theory is formulated to meet the quite different aim of securing the objectivity of scientific concepts.

Revisionists argue that the central problem motivating Carnap's reductionist project is not the empiricist problem of how we can have knowledge of the external world, but the more general Kantian problem of how objective knowledge is possible. Carnap raises the problem as follows:

As suggested by Richardson's (Reference Richardson, Fine, Forbes and Wessels1990) paper, “How Not to Russell Carnap's Aufbau,” the foundationalist reading has gained its legacy due to readers' tendencies to read Russell's ‘external world program’ directly into the Aufbau. Although Carnap does take Russell as a fundamental point of inspiration, what Carnap inherits from Russell is his logic, not his problem. In Carnap's hands, Russell's logic is not applied to justify our knowledge of the external world, but rather to articulate a new conception of objectivity that assimilates logicist and neo-Kantian principles.Footnote ⁵

3. Logic, Structure, and Objectivity

According to revisionists, the aim of construction theory is to secure the objectivity of scientific concepts by capturing what is intersubjectively communicable in conceptual knowledge, viz., its structural form (Friedman Reference Friedman1999, chap. 5; Richardson Reference Richardson1998, 180–206).Footnote ⁶ For Carnap, the objectivity of conceptual knowledge lies in the structural properties of relations obtaining between concepts (§§ 6, 10, 12, 16). Whatever is not structural (i.e., whatever is empirically ostensible) in a relation is subjective, and must be removed in construction theory. Carnap explains:

Here, Carnap argues that to secure the objective meaning of concepts, it is necessary to provide a formal description of the relations obtaining between concepts. The logic of Principia Mathematica, conceived as a conventional tool for formal analysis, serves this role in construction theory. Carnap writes: “Logic (including mathematics) consists solely of conventions concerning the use of symbols, and of tautologies on the basis of these conventions” (§ 107, emphasis in original).

More precisely, Carnap describes the structural properties of relations with the definitional tool of ‘purely structural definite description’ (§§ 11–15, hereafter PSDD). A PSDD uniquely specifies a concept by defining a concept exclusively in terms of relational structure without reference to the content of the relation or its relata (§ 11). The essential feature of a PSDD is that it is capable of capturing the minimal properties necessary to unequivocally specify relevant concepts (§ 13). Carnap presents an example of how individual objects on a map of a railroad network can be specified without reference to ostensive objects (§ 14). The idea, roughly, is that any of the relevant objects (i.e., train stations) can be specified merely with reference to a single relation (‘nextness’) via PSDDs (see Richardson Reference Richardson1998, 47–49). In the context of the constructional system presented in the Aufbau, Carnap's strategy is analogous. All concepts are to be reduced to purely formal definitions that refer to structural properties obtaining between concepts without any reference to ostensible content.Footnote ⁸

On the revisionist reading, construction theory aims to provide an objective account of what scientific concepts mean by rationally reconstructing them into an epistemological genealogy. Logical definitions (constructional definitions) are conventional introductions of signs that redefine concepts in terms of previously defined or primitive concepts (Richardson Reference Richardson1998, 185–187). Carnap's proposal assumes both a structuralist conception of objectivity and the possibility of the ‘unity of the object domain’ (§ 4).Footnote ⁹ The nature of Carnap's solution is revealed when he writes:

The idea is that objective conceptual meaning in science is purely structural. Since all concepts, in principle, belong to the same conceptual domain, the objective meaning of all scientific concepts can be expressed in a language that is sufficiently rich to capture the essential structural properties of concepts. The resources of the logic provided by Whitehead and Russell (Reference Whitehead and Russell1910–1913) meet Carnap's desideratum nicely.

4. Rational Reconstruction as a Method of Justification

The aim of the remainder of this paper is to articulate how Carnap's method of rational reconstruction is a form of epistemic justification. In doing so, I hope to provide a more balanced perspective on the nature of Carnap's epistemological project in the Aufbau. Although revisionists are correct to stress that construction theory is motivated to secure the objectivity of scientific concepts, they fail to sufficiently acknowledge that this goal facilitates a more general goal within the context of the epistemology of the Aufbau, viz., to provide an empiricist method for the justification of scientific conceptual knowledge. In drawing attention to this more general goal of construction theory, I hope to clarify what is both correct and obscured in the revisionist account on the nature of the epistemology of the Aufbau. In particular, I argue that when the goal of construction theory is recast from objectivity to justification, what is suggested is not an exclusively neo-Kantian project, but a project that assimilates aspects of both neo-Kantian and empiricist epistemology.

In the preface to the first edition of the Aufbau, Carnap indicates that construction theory is motivated to address questions of epistemic justification by means of reductionistic methodology:

Now, if construction theory is a method of epistemic justification, then how does Carnap's project facilitate this end? A succinct answer can be found in the opening lines of Pseudoproblems in Philosophy (hereafter, Pseudoproblems), Carnap's ([1928] Reference Carnap, George and Carnap1967b) non-technical article published in the same year as the Aufbau:

This passage, I think, encapsulates the nature of Carnap's method of construction theory in the Aufbau. Footnote ¹⁰ Consider an Aufbau-style rational reconstruction. Construction theory exhibits how the meaning of concepts at higher levels are related to the meaning of concepts at lower levels. In clarifying the inferential and non-inferential (or more accurately, ‘presumed to be valid’) parts of knowledge and showing how they are related, a constructional system addresses questions of justification by specifying clearly on what basis knowledge is inferred. This sort of formal clarification amounts to a ‘rational justification’ of concepts insofar as it provides a demonstration of how such concepts can be intersubjectively understood. Carnap's understanding of justification as being relative to a basis of ‘presumed validity’ is strictly opposed to the ideal of attaining absolute justification—his counter-suggestion is that justification can only be secured relative to the choice of a constructional system.

To some extent, the general notion of justification sketched above can be assimilated within the revisionist view of construction theory. What is suggested by the revisionist view is that insofar as Carnap's project addresses questions of justification, it falls squarely in the neo-Kantian tradition (see Friedman Reference Friedman1999, 124–132). In contrast with empiricist epistemology, which takes the justification of beliefs, the refutation of skepticism, and the provision of a certain foundation for knowledge in an empirical basis as its primary problems, neo-Kantian epistemology is engaged in the prior problem of how objective judgements are possible. This neo-Kantian problem is essentially a question concerning what makes judgments capable of being either true or false. From this perspective, questions of epistemic justification follow from considerations of how subjective ideas come to acquire objective meaning. According to revisionists, Carnap's answer is conventionalist: objective judgments are possible by means of placing concepts into a global constructional system with a conventionally chosen basis. The significance of rationally reconstructing concepts in this way is that it stipulates objectively meaningful conditions, within that system, for what makes statements about a concept true or false. Carnap writes: “[T]he indication of the nominatum [read, meaning] of the sign of an object, consists in an indication of the truth criteria for those sentences in which the sign of the object can occur” (§ 161).Footnote ¹¹ In other words, attaining conceptual objectivity through rational reconstruction results in definitive answers to questions posed about concepts within that system. Although I believe that this picture of justification is, at the very least, implicit in the revisionists' account, their failure to sufficiently articulate this particular aspect of construction theory has consequently obscured the extent to which Carnap's method can be located in the empiricist tradition. For Carnap, the ultimate significance of securing the objectivity of concepts is to ensure the empirical testability and answerability of concepts. As such, construction theory is an empiricist method of justification in the sense that it aims to rationally justify concepts by showing how questions posed about them can be empirically answerable.

5. Rational Justification, Science, and Empiricism

In the concluding chapter of the Aufbau, under the heading “Aims and Limits of Science,” Carnap discusses scientific methodology with reference to the constructional system outlined in his book (§§ 179–183). This discussion elucidates the precise sense in which construction theory serves as an empiricist method of justification. Carnap understands science as a ‘system of conceptual knowledge,’ and he maintains that its aim is to ‘find and order true propositions’ (§§ 179–180). This aim can be reached through two interrelated tasks: (1) the introduction of concepts in a constructional system, and (2) the ascertainment of the empirical relations between the concepts of a constructional system (§ 179). Hence, Carnap maintains that science aims two meet two tasks, the first conventional and the second empirical, but both equally important.

Carnap's discussion of science suggests that construction theory functions as a method of justification by relating (or coordinating) the conventional and empirical parts of conceptual knowledge. The first task of science (in a logical rather than temporal sense)—rational reconstruction—demands that scientific knowledge be formulated in an objective constructional system because “to be able to make statements about objects at all, we must be able to construct these objects” (§ 179). The significance of this requirement, for Carnap, is that it raises the very possibility of the second task of science, “the investigation of the nonconstructional properties and relations of the objects” (§ 179). In this sense, a constructional system is constitutive of scientific knowledge insofar as the reconstruction of concepts in a constructional system yields objectively decidable and answerable empirical judgments and tests about concepts (§ 180).Footnote ¹² Carnap writes:

Thus, we can see how construction theory serves as an empiricist method of justification. For Carnap, it is only against the background of an antecedently stipulated constructional system (e.g., one with an autopsychological basis) that objectively meaningful empirical questions about concepts can be raised. Only after such a reconstruction has taken place do the concepts in the system acquire objective meaningfulness in the sense that statements that are made about reconstructed concepts will be subject to empirical testing. In this manner, concepts are rationally justified insofar as reconstruction gives rise to conditions of empirical answerability for statements made about concepts.

The understanding of construction theory as an empiricist method of justification makes sense of Carnap's claim that science ‘has no limits,’ meaning that ‘there is no question in science whose answer is unattainable’ (§ 180). Carnap writes:

For Carnap, science is unlimited in the sense that the truth or falsity of questions posed about reconstructed scientific concepts can (in principle) be decisively addressed by empirical means. As such, reconstructed concepts are ‘rationally justified’ in the sense that questions posed about them are empirically ascertainable.

In construction theory, Carnap's usage of ‘empirical theorems’ (§ 106), in building constructional systems, ensures the empirical answerability of questions posed about reconstructed concepts. Carnap maintains that there are only two types of theorems that are utilized to build any constructional system: (1) analytic theorems, which are deduced exclusively from the constructional definitions, and (2) empirical theorems, which specify the relations between constructed concepts that are only ascertainable through experience. Carnap writes: “If an analytic theorem is transformed into a statement about the basic relation(s) of a system, a tautology results; if an empirical theorem is thus transformed, it indicates empirical, formal properties of the basic relation(s)” (§ 106). Hence, empirical theorems have the peculiar status of being both formal and empirical. They are formal insofar as they are purely structural statements, however, they are empirical insofar as they express relations that are only known through experience (Richardson Reference Richardson1998, 66). In the context of Carnap's ‘analysis of reality,’ “this means that … empirical theorems express an empirically ascertained state of affairs” (§ 106). That is, in ‘realistic language,’ the mode of speech of the empirical sciences (§§ 95, 98), these theorems function to determine the empirical truth conditions for questions posed about the concepts of a constructional system (it should be emphasized that these truth conditions refer to an empirical or constructional concept of reality). By building a constructional system with empirical theorems, which serve to set the limits of empirical discernibility with respect to reconstructed concepts (see Hudson Reference Hudson1994, 240–242), Carnap ensures that questions made about these concepts will be empirically ascertainable.

The analysis above suggests that revisionists fail to adequately address the role of empiricism in construction theory. In focusing their analyses to establishing the claim that construction theory aims to secure the objectivity or intersubjectivity of conceptual knowledge, revisionists have failed to sufficiently address the question of why reconstructing concepts in an objective system is important. As suggested here, the significance of intersubjectivity, for Carnap, lies precisely in the fact that questions posed about reconstructed concepts will be empirically ascertainable. Thus, construction theory is a characteristically empiricist method insofar as it recommends that concepts are to be justified by showing their connection with empirical experience (cf. Carnap Reference Carnap1937, 33; Hudson Reference Hudson1994, 239–242; Uebel Reference Uebel1996, 283–289). On this reading, construction theory is a method of justification that requires the relationship between reconstructed concepts and observable experiences to be formally demonstrated (and the clarification of this particular relationship is what ultimately provides the ‘rational justification’ of concepts).

It is important to emphasize that Carnap's empiricist requirement of answerability is not limited to a constructional system with an autopsychological basis, but applicable to all constructional systems. In considering alternative bases for constructional systems (§§ 55–64), Carnap is most optimistic towards a physical basis, citing the advantage that “there is a greater degree of intersubjective agreement” (viii). To see that intersubjectivity is not simply a virtue in itself but a means to empirical answerability, consider the possible physical bases that Carnap proposes, viz., ‘electrons,’ ‘four-dimensional space-time points,’ or ‘world lines’ (§ 62). For Carnap the important feature of each of these intersubjective bases is that: “The perceptible physical things and properties can then easily be constructed from the things and properties of physical science, since they are uniquely determined by them” (§ 62, cf. Carnap Reference Carnap1936a, 463–468; Reference Carnap and Smeaton1937, 9–14). This particular point is clarified by Carnap (Reference Carnap1963b) in a reply directed towards Goodman:

As this passage makes clear, the importance of intersubjectivity, for Carnap, is that it permits agreement among various observers on the ‘observable properties’ of concepts in the system, i.e., it stipulates empirically answerable truth criteria for statements made about concepts. The fact that all constructional systems are subject to an empiricist constraint of this kind suggests that construction theory can legitimately be characterized as an empiricist method of justification.Reference Schilpp

6. The Nature of Justification in the Aufbau.

Although I have argued that construction theory can be characterized as an ‘empiricist’ method of justification, this should not underscore the fact that Carnap takes himself to be redefining ‘empirical justification’ as a purely formal notion. The justification of concepts, for Carnap, is achieved by exhibiting, in the language of logic, both the epistemological basis and the truth conditions for objective judgments about concepts. This exhibition or logical analysis of meaningfulness amounts to showing how the logical relations between a concept and other concepts at different levels relate to the empirical answerability of such concepts. As such, construction theory is a method that redefines ‘empirical justification’ as a formal-logical relation obtaining between groups of statements (cf. Uebel Reference Uebel1992). In addition to providing Carnap with the formal means to frame scientific concepts objectively within particular constructional systems (thereby providing a rational justification for them) logic also plays a prominent meta-philosophical role in the Aufbau. In this latter role, logic provides Carnap with the means to recast all of traditional epistemology that is infected with metaphysical assumptions (Richardson Reference Richardson1998, chap. 8).

Logic grounds Carnap's metaphysical-neutrality in the Aufbau with a purely structural language applicable to all aspects of epistemology, limiting discourse to formal definitions that neither require nor permit any ‘deeper interpretation.’ For Carnap, construction theory represents the metaphysically-neutral standpoint that can capture the core agreement of all epistemological schools. Carnap presents this common point of agreement as follows:

By reconstructing this picture of knowledge into purely logical terms, Carnap takes himself to be withdrawing from the fruitless debates that occur at the level of metaphysics. On this understanding, logic provides Carnap with the a priori tools necessary to replace metaphysically based epistemologies with a formal-logical discipline (Friedman Reference Friedman1999, 94; Richardson Reference Richardson1998, 26–28).Footnote ¹³ As expressed in Carnap's intellectual autobiography, Russell's (Reference Russell1914) conception of ‘the logical-analytic method of philosophy’ is the main inspiration driving the centrality of logic in the Aufbau (see Carnap Reference Carnap1963a, 13–20). The most explicit endorsement of this meta-philosophical role for logic is stated poignantly by Carnap ([1928] Reference Carnap, George and Carnap1967b) in Pseudo-problems: Footnote ¹⁴

This passage indicates the centrality of logic in Carnap's conception of epistemology. Logic plays a fundamental role not only in driving Carnap's metaphysical-neutrality, but in clarifying the very aims of philosophy. In the Aufbau, logic grounds the whole of project of epistemology by recasting the principles of epistemology in logical terms (Richardson Reference Richardson1998, 191–196).

In the Aufbau, ‘justification’ consists of a logical analysis of conceptual knowledge. To justify a concept is to reconstruct it as a constructional definition located within an ordered constructional system in which concepts are reducible to a basis (e.g., a sense-data or physical basis) that is presumed to be valid. Rational reconstruction serves as a method of justification insofar as it shows clearly on what basis conceptual knowledge is presumed to be objectively and empirically meaningful. Unlike traditional empiricist notions of justification (in the tradition of Locke, Hume, and Russell) that aim to justify our knowledge of an ‘external world’ in an empirical foundation of certainty, Carnap's notion of justification aims to rationally justify conceptual knowledge by showing how conceptual knowledge can be objectively meaningful.Footnote ¹⁵ Objective meaningfulness, moreover, gains its significance only insofar as it clarifies the empirical basis of concepts. As Carnap stresses repeatedly, the object of reality to be reconstructed in construction theory is a ‘constructional’ or ‘empirical’ concept of reality (§§170–178). As such, Carnap's method rejects traditional absolute notions of justification in favor of a more deflationary notion that is explicitly conceived as being relative to the choice of a constructional system. Thomas Uebel (Reference Uebel1996) characterizes Carnap's strategy nicely: “The claim to scientific knowledge is not justified from without, rather, it is being shown what justification is in science and how objectivity best be thought of from within science” (396, emphasis in original).

The notion of justification that I have argued Carnap is forwarding in the Aufbau anticipates ideas that are crystallized in Carnap's mature ‘philosophy of linguistic frameworks’ culminating in his classic paper “Empiricism, Semantics, and Ontology” ([1950] Reference Carnap1956). Carnap's distinction in this work, between ‘internal’ and ‘external’ questions, gives rise to a related distinction between ‘theoretical’ and ‘pragmatic’ justification. In my view, Carnap clarifies, in this later work, the status of questions regarding the choice of a basis for a constructional system as external questions (i.e., questions that are not meaningfully answerable as true or false) that only admit pragmatic justification. By contrast, questions regarding concepts that are reconstructed within a constructional system are internal questions (i.e., questions that are meaningfully answerable given the rules of an agreed upon constructional system) that admit theoretical justification. Like the method of justification advanced in the Aufbau, all notions of truth or falsity, and hence theoretical justification, of internal assertions are understood by Carnap as being relative to the choice of a constructional system. Moreover, Carnap's conventionalist notion of justification in his later writings also prescribes objective meaningfulness (i.e., intersubjectivity) as the key requirement for theoretical justification.

7. Conclusion

In this paper, I argued that the epistemological project of the Aufbau, construction theory, is a method of justification. I articulated this argument with reference to the neo-Kantian interpretation of the epistemology of the Aufbau defended by Friedman and Richardson. While I agreed with revisionists in their general claim that construction theory is motivated to secure the intersubjectivity or objectivity of scientific concepts, I argued that this project is couched within a broader goal of providing a method for the ‘rational justification’ of concepts. I argued that construction theory meets this broader goal in its proffered method for formally reconstructing concepts. By rationally reconstructing concepts into an objective genealogy, reconstructed concepts are rationally justified because their empirical conditions of answerability or decidability are specified. On this understanding, the rational justification of concepts, for Carnap, involves both a conventional and empirical aspect. On the one hand, justification is conventional insofar as any constructional system is the result of a conventional or pragmatic decision regarding the basis of the reconstructed system. On the other hand, justification is empirical because after the basis of a constructional system is chosen, objective empirical conditions of answerability are specified. What this specification of empirical conditions amounts to is the provision of truth conditions for questions posed about concepts within a system. Because the answers to such questions are relative to a chosen basis, Carnap's notion of justification is conceived as being relative to the validity of the chosen basis of a constructional system.

In shifting the focus of analysis of Carnap's project from issues of objectivity to justification, I argued that the nature of Carnap's epistemological project is more accurately characterized as involving both neo-Kantian and empiricist aspects, rather than falling neatly into either tradition. The fact that Carnap's perspective in the Aufbau is meant to be reconciliatory, is indicated in the preface to the second edition:

Carnap's characteristic response to this perceived overemphasis on opposing sides is to synthesize both views in the neutrality of logic. If the motivating epistemological insight of construction theory is shifted—from neo-Kantianism, per se—to such a synthesis, then what is correct and obscured by both the received and revisionist views can be stated with clarity. The received view is correct in its supposition that Carnap's project is addressing questions of justification by adopting reductionistic methodology, however, it obscures this project by supposing that the search for justification implies the search for certainty. The revisionist view is correct in its supposition that Carnap's project is addressing questions of objectivity, however, it obscures this project by supposing that the attempt to construct objective meaning has no significant connection with the aim of empirical justification. As urged in this paper, a more comprehensive understanding of construction theory can be gained by focusing attention on the issue of justification. In defending such a reading, my aim was to bring attention to a theme that I believe is already implied by the revisionist interpretation but, in my view, not sufficiently articulated. In elaborating this theme, I hope to have made a contribution—building upon the significant steps that have been already made by revisionists—to a better understanding of the epistemology in Carnap's Aufbau.