TY - JOUR
AU1 - Olmos,, Paula
AB - Abstract This paper makes use of the concepts and theoretical framework developed within the field of Argumentation Theory to account for the structure and characteristics of abduction and of the comparative processes of weighing explanatory hypothesis. It elaborates an analysis of abduction based on its consideration as a meta-explanatory argumentation scheme while elucidating its relations with abductive reasoning and inference. The conceptualization of comparative processes of weighing explanatory hypothesis as complex and varied argumentative structures is presented as an alternative to the idea of providing a formally rigid and supposedly universal account of ‘inference to the best explanation’. 1. Introduction This paper makes use of the concepts and theoretical framework developed within the field of Argumentation Theory to account for the structure and characteristics of abduction and of the comparative processes of weighing explanatory hypothesis. Argumentative accounts of abduction have not been so far as frequent as could be expected. This much is admitted by Wagemans [23], who has recently published some papers on abduction using a pragma-dialectical framework. Yu and Zenker [28] have also lately paid some attention to abduction following a Peircian inspiration and applying the methodology of argumentation schemes and critical questions. But hitherto the most extensive treatment of abduction from the ranks of argumentation scholars is due to Walton who, in 2004, published the monographic work Abductive Reasoning. Walton’s approach to abduction, based on the methodology of ‘dialogue types’ and models that he developed together with Krabbe [25] contributed some interesting ideas, as a proposed dialogical model of explanation [24, 79ss, cf. 17], but builds on somewhat confusing starting points. In particular, I would say that Walton’s account does not clearly trace or at least clearly exploit three distinctions that are the basis of my own argumentative proposal: (a) the discrimination between mental processes of reasoning and communicative exchanges of giving reasons; (b) a more careful distinction between the communicative acts of explaining (i.e. giving explanatory reasons) and communicative acts of arguing (i.e. giving justificatory reasons); and (c) an analysis of simple abduction as a distinctive argumentation scheme based on a characteristic kind of reason-for-a-conclusion—and, thus, a characteristic kind of warrant in Toulmin’s [21] sense—as opposed to what I will call inference to the best explanation (IBE)-type exchanges: complex and varied (typically dialogical and intrinsically dialectical) communicative processes of selecting, assessing and weighing alternative explanatory hypotheses, purporting the use of varied argumentative, counter-argumentative and meta-argumentative, both schemes and structures. I will develop and support separately the two first distinctions (Sections 2 and 3) and use them in construing an argumentative (meta-explanatory) scheme for abduction (Section 4). This will allow me to differentiate between such a relatively simple argumentative scheme and an organic and meta-argumentative account of IBE-type exchanges in terms of a variety of tools for argument analysis (Section 5). In my proposed argumentative approach to abduction, I will make use of a Toulminian argumentation scheme, as opposed to the ‘Premises-plus-Conclusion Model’ used by both Walton and Yu and Zenker. Toulmin’s model of argument [21] is mainly characterized by the distinction between the data or grounds that amount for the justificatory reason supporting the conclusion (or claim) of an argument and the underlying principle (acting as inference license) that makes of those data or grounds a reason for such a claim and specifies what particular type of reason they constitute. Such principle1 is called the argument’s warrant. In my view, it is precisely this distinction and the awareness of the crucial role of the substantive kind of relations expressed by argumentative warrants in construing and understanding arguments that make Toulmin’s proposal depart radically from any formal model of argument based on analytic and merely semantic relations. I will use the following kind of diagram (developed by [10]) to represent arguments according to Toulmin’s distinctions: $$\begin{equation*} \begin{array}{ccccc} & & & & \boxed{\qquad\quad\textrm{Data}\ \ \qquad\quad}\\ \textrm{Warrant:} & & & & {\textbf{So}}\\ & & & & \boxed{\qquad\quad\textrm{Claim}\qquad\quad}\end{array} \end{equation*}$$ The use of Toulmin’s model, moreover, does not preclude the application of what constitutes the most useful instrument of the methodology of argumentative schemes, i.e. the identification of the critical questions associated with a particular argument type. Quite the opposite: Toulmin’s distinctions help guiding and ordering different sets of critical questions and prevent the enunciation of too formalistic and unspecific ones such as ‘are the premises true?’ or ‘does the conclusion follow from the premises?’ that are usually misguided in argumentative settings. However, what is missing in Toulmin’s model, representing a single or atomic argument, are precisely adequate tools to describe and interpret inter-argumentative relations. In a series of papers,2 Marraud [8] has developed and systematized tools which account for a variety of argumentative (co-oriented), counter-argumentative and meta-argumentative structures that conspicuously appear in communicative exchanges and that constitute the basis of his ‘arguments’ dialectic’. In our case, these tools become precious to describe and account for IBE-type exchanges (see Section 5). 2. Abductive reasoning vs. abductive argument Regarding distinction (a) among those mentioned in the Introduction, although there are in fact communicative uses of the term reasoning (e.g. ‘interpersonal reasoning’), the distinction between mental processes of reasoning and communicative exchanges of giving reasons is widely admitted and advocated as significant in philosophical literature, especially within the field of argumentation studies. However, at least in my opinion, it has not been so far conveniently and fruitfully explored in the explication of concrete problems. I take the distinction between mental reasoning and communicative arguing to constitute a step further along the line of Woods’ [26] discrimination between the following: (i) the consequences a group of premises may have within a logical space (the space of logical derivation), for whose delimitation, there is no need to count on the involvement of any agent: ‘There are no heads in logical space’ [26] and (ii) the epistemic consequences that someone may, within her own psychological space (the space of reasoning and inferring), first identify or spot (in a recognition subspace) and then draw or infer (in an inference subspace).3 Psychological space is not only different from logical space but also Woods claims that the latter does not provide the normative counterpart for the former. But, unlike Woods, who explicitly dismisses this possibility, here it is claimed that an exploration and better understanding of the distinctively public space of reason-giving and arguing has much to offer both from a descriptive and a normative point of view. In such public, communicative and interactive space we may talk about the following: (iii) the (also epistemic and agent-related) consequences that someone may present to others as justified by considerations adduced in the same communicative context as reasons in its favor. In the case of abduction, the idea of this move is basically switching the current emphasis on accounting for how individuals mentally make use of more or less intuitive constrictions when suggesting plausible—and more or less creative—explanatory hypotheses, to the interactive, interpersonal space of collective discussion from which (I claim) such internalized constrictions really emerge. This idea agrees with recent cognitive approaches of a pragmatist descent like Mercier and Sperber’s [12, 13] (genealogical and evolutionary) ‘argumentative theory of reasoning’. Focusing on communicative argument over mental reasoning favors attention to the socialized processes through which we learn to recognize and use conventional argumentative patterns4 and emphasizes the intrinsically normative nature (from an insider’s point of view) of the communicative practices of asking for, giving and assessing reason(s). If we approach abduction starting from its communicative counterpart, offered in a (typically) dialogical justificatory exchange, instead of referring to its mental (misleadingly instantaneous) model, we may transfer all additional requirements made to our justificatory claims to the continuity of argumentation and argumentative interactive assessment, i.e. to the ‘continuum between argument and argument criticism’, mentioned by Pinto [18]. Thus, we may avoid somewhat obscure notions as Peirce’s ‘intuitive flash’, ‘insight’ (CP 5.173), ‘power of guessing right’ (CP 6.530) as sources for abductive conclusions and examine, instead, the substantive criteria for argument assessment that the context and the particular argumentative practice in which the agents are involved would have construed along their dialogical and argumentative history.5 Finally—although Woods’ [26] account of defeasible and non-monotonic relations of consequence and their virtuous fragility does, in my opinion, already do the job pretty well—I think it is still important to emphasize the substantive and ampliative character of any argument worth discussing, as based on reasons and substantive (i.e. epistemic, in Woods’ terms) warrants and not on formal relations of implication, which only hold when the conclusion’s contents are already included in the premises. In the case of abduction, which has been considered by many as the paradigm of creative and ampliative reasoning—something that puzzles and bewilders logicians and becomes a source of fascination for many in philosophy and cognitive science—the move toward the public exposure of the rational criteria6 used for its acceptance, rejection, discussion or limitation might be especially beneficial. Therefore, our focus remains on abductive argument and not on abductive reasoning, acknowledging nevertheless that the link between both might be worth examining in further research. According to this framework, presenting in a communicative interchange an abductive argument is to support an in principle theoretical or factual claim (typically mentioning either unobservable or merely unobserved entities, properties and processes) on the basis of shared data (typically observable, well-known, taken for granted or assumed) precisely because it provides some conceivable explanation to them. The basic elements of an abductive argument—using, as has been previously announced, Toulmin’s model—are the following: Conclusion/Claim: an ‘explanatory hypothesis’ H, usually presented as a factual statement, although, depending on the requirements of the context it may be easily reinterpreted as a practical or even evaluative conclusion of the kind: ‘we should explore hypothesis H’, ‘Hypothesis H is worth exploring’. Reasons/Data: usually empirical, observable but in any case presented as shared or agreed upon data, and nevertheless surprising data, i.e. data requiring explanation (what makes of them a potential explanandum). However, this preparatory condition might be contested in an ensuing discussion. Warrant: what makes of the data a justificatory reason for the conclusion (the hypothesis) is that such hypothesis could explain them. These elements may be represented thus as follows: $$\begin{equation*} \begin{array}{cccc} & & \boxed{\ \ \quad{\textbf{Data}: Shared (usually empirical) data}\ \quad}\\ {\textbf{Warrant}: Hypothesis \textbf{could explain} data} & & {\textbf{So}}\\ & & \boxed{\ \quad\quad \qquad\quad{\textbf{Claim}: Hypothesis}\ \qquad\quad\quad\quad}\end{array} \end{equation*}$$ In the case of abductive arguments, it seems as though the kind of warrant I propose is not exactly ‘of a more general scope’ than the data and claim of the argument, referring exactly to the same items as its basic elements. Nevertheless, this is just apparently so as such warrant introduces between these items a substantive kind of relation, namely ‘explanation’, which responds as well (as will become clear in Section 3) to a variety of principles on which it may be based. The concretion of the type of explanans that the Hypothesis is vis-à-vis the Data (taken as explanandum) provides the degree of principled generality that an argument requires to be so recognized and understood (and eventually assessed) by an interlocutor. Taking in account that abductive arguing is (as is abductive reasoning and inferring, precisely for that matter) firmly and extensively incorporated in our daily cognitive life, the move from data (D) to hypothesis (H) is in most occasions rather directly expressed and communicated, as in the following example: - There is no yoghourt left in the fridge, so Tom finally had the last one for dinner. If the interlocutor would ask ‘why do you say so?’ or ‘where do you get that from?’, we could in principle think that she has not understood the proposed connection between data and hypothesis and answer by making the warrant explicit: - Well, that would surely explain it … In the case of abductive arguments (and this does not necessarily happen with any kind of argument) this seems somewhat artificial, precisely because of their pervasiveness and basic nature. The usual thing is to automatically recognize the proposed link (the way the data may constitute a reason for the conclusion) and go directly to critical assessment, usually, if the argument and its claim are not directly accepted, in the form OF diverse kinds of counter-arguments: objections to data, rebuttals directed to the warrant (or its application) or refutations of the claim based on alternative reasons [11]: [Objection]: - Look again, I think there’s one left … [data were not shared, after all, as the speaker initially assumed]. [Rebuttal]: - But there were 10 yoghourts yesterday … [even if Tom took one for dinner, this does not explain that there is none left, after all]. [Refutation]: - You are right, there is no yoghourt left in the fridge, so maybe it is somewhere out there getting warmer … [if there are alternative explanatory hypotheses available, we will have to compare them, look for the best (available) one. So, if it matters, the enquiry will continue …] Before arriving at the assessment stage, though, we need to take into account the distinction between acts of arguing and acts of explaining (distinction (b) of the Introduction) that will allow us to complete the description of the basic or atomic but, ultimately, not-so-simple structure of the abductive argument. 3. Arguing vs. explaining The distinction between acts of arguing and acts of explaining is, in principle, better established and widely recognized in argumentation theory (and in general philosophy for that matter). (a) Acts of arguing involve offering reasons in order to justify a content (the claim) that is contextually deemed doubtful (or yet unjustified), seeking the interlocutor’s (or wider audience’s) persuasion about its correctness. (b) Acts of explaining also involve offering reasons but this time in order to make comprehensible an already accepted content (the explanandum), seeking the interlocutor’s understanding about its (causal) origin, its relation with other phenomena or its overall congruity and dispelling wonder or surprise regarding it. According to this characterization, explaining and arguing would be structurally similar—in both cases reasons are given as related to a content (a claim or an explanandum)—and pragmatically different—the status of that content and the pragmatic aims of the two activities are different (cf. Marraud 2013, p. 24–26). This structural similarity allows us to build a model of explanation that mirrors Toulmin’s model of argument. As I have already said, in Toulmin’s model of argument the reasons for a claim (identified as the particular case’s data) are presented as reasons precisely since complying with the requirements of a certain inference license or principle—the argument’s warrant. Likewise, in my proposed Toulminian model of explanation, the explanatory power of the reasons offered in the explanans is grounded upon certain idea or principle of what counts as explanatory in the context in which the explanation is presented. The kind of diagram used to represent explanations will be as follows (notice that the conventional connective expressing the link is a different one): $$\begin{equation*} \begin{array}{ccccc} & & & & \boxed{\ \ \qquad\quad{Explanans}\ \ \ \qquad\quad}\\ \textrm{Explanatory principle:} & & & & {\textbf{That's why}}\\ & & & & \boxed{\qquad\quad\textrm{Explanandum}\qquad\quad}\end{array} \end{equation*}$$ In the case of abduction, Walton’s monography [24, Chap. 3, pp. 97–121] does indeed already acknowledge this distinction, offering first a dialogical model of explanation (pp. 79ss) and then a second dialogue model (pp. 240–243) that portraits an interactive justificatory process of the best explanatory hypothesis, which he considers as equivalent to abduction. The distinction is also present in the work of philosophers who discuss kinds of reasons [1]. However, the intrinsic connection of these two different acts that is characteristic of abduction, a kind of justificatory argument in which what is justified is precisely an explanatory hypothesis and more precisely for its explanatory power, forces us to be extra careful in this case, when modeling and understanding abduction. The analysis offered in this paper of abduction (of simple abduction) as an argumentative meta-explanatory scheme tries to cope with this difficulty. 4. A meta-explanatory model of abduction As we have seen, our Toulminian scheme for abduction identifies as the abductive arguments’ warrant the possibility of construing an explanation of the data taken as explanandum based on the hypothesis used as explanans. So, if the basic diagram for an abductive argument is $$\begin{equation*} \begin{array}{cccc} & & \boxed{\ \ \quad{\textbf{Data}: Shared (usually empirical) data}\ \quad}\\ {\textbf{Warrant}: Hypothesis \textit{could explain} data} & & {\textbf{So}}\\ & & \boxed{\ \quad\quad \qquad\quad{\textbf{Claim}: Hypothesis}\ \qquad\quad\quad\quad}\end{array} \end{equation*}$$ it is possible to construe an ‘expanded diagram’ including the details of the related explanation—an explanation that is not exactly given in the abductive argumentative act but just, mentioned or alluded to as what makes of the empirical data a justificatory reason for the (theoretical) hypothesis—thus, Notice that this does not constitute an act of explaining but an act of arguing that urges the discussion of its conclusion (the ‘explanatory hypothesis’). In any case, this expanded, explicitly meta-explanatory diagram may help us locate and identify different possible targets of the issuing assessing discussion urged by an abductive proposal. Presenting an abductive argument means, thus, to offer at least a minimally justifying reason for a hypothesis (which is a doubtful content, what is at stake) adducing the factive quality of the data, for which the hypothesis is deemed to be explanatory.7 This means that it is suggested that the hypothesis could be used (in a different act and setting, allegedly once accepted) as an explanans or explanatory reason for the acknowledged data. 5. Abduction vs. IBE In my view, a simple abductive argument is, thus, completely described. It constitutes a characteristic type of argument, using a characteristic kind of reason (what means a characteristic kind of warrant). However, as any other argument, it is liable to a subsequent discussion in which further clarifications, further justificatory reasons and requirements may be demanded (… or not, depending on the context) to improve the epistemic status or acceptability of its conclusion, if and only to the extent that this is our goal (e.g. characteristically in Science). In argumentation theory, such discussions are usually modeled by means of the battery of ‘critical questions’ typically attached to an argumentative scheme. In the case of abduction, some of these questions (not necessarily all of them) involve the comparison of the claimed hypothesis with other alternative explanatory hypotheses. So typically (thus not necessarily), when the context allows for the proposal of different explanatory hypotheses and the agents involved in the particular situation assume the necessity (or the institutional requirement) to choose one among them as the ‘best (available) explanation’, a rather more complex justificatory interchange starts. This usually involves the comparative assessment of the first arguments advanced in favor of each hypothesis (some of them, although this is not a necessary condition, possibly abductive) and it may easily require further arguments and meta-arguments in support of the final conclusion (i.e. the chosen hypothesis). This brings us back to distinction (c) of the three mentioned in the Introduction. It is easy to see now how the also argumentative but much more complex process of comparison, weighing and selection of the best available one among several explanatory hypotheses departs from and cannot be squeezed into our proposed model. In the divide established in the philosophical literature between those who assimilate abduction to IBE [3, 24] and those who maintain the distinction [6, 7, 28], I definitely side with the latter, although I contend that universal and formal models of IBE say really not much about the way IBE-type discussions are conducted. I further claim that at least certain strands in argumentation theory offer us, instead, a number of analytic tools that may help us model (and better understand) such concrete IBE-type discussions, i.e. argumentative interchanges aiming at justifying a best available explanation as such. The idea is to reconstruct the very varied ways some typical clauses appearing in certain models of abduction or IBE (e.g. ‘H1 explains data better that the other available hypotheses’ or ‘H1 complies with criteria S1 … Sn’) in turn come to be justified. These tools aim at modeling the argumentative structures that reflect inter-argumentative and meta-argumentative relations, of which the most complex and sophisticated account for the comparative weighing or hefting of reasons, which is typical of such kind of discussions.8 They include the following: (a) Argumentative co-oriented structures (combinations of reasons): including co-oriented reasons in conjunction (i.e. linked arguments) and co-oriented reasons in disjunction (i.e. convergent arguments).9 (b) Counter-argumentative structures: objections to premises, rebuttals of links (affecting the warrant’s validity or applicability) and refutations of conclusions (counter-arguments supporting opposing conclusions). (c) Structural meta-argumentative weighing: i.e. justified comparative assessment of arguments.10 All these structures typically appear in IBE-type discussions. So, if the tremendous variety of such exchanges is the bad news about them for philosophers, the good news is their describability in terms of this battery of distinctions. 6. Argumentative analysis of an example of IBE The following paragraph appears in John Herschel’s Preliminary Discourse on the Study of Natural Philosophy (1851, pp. 144–145). It is a discussion, presented by a single author, regarding the reasons that favor one of the explanations given for a certain phenomenon as the best among them, in scientific terms. It is easy to see that the presentation reconstructs a discussion that has taken place for many years among natural philosophers and (lately) geologists. I have divided it into five parts in order to refer to them in the subsequent (just outlined) analysis.11 [Part 1.] The phenomenon of shells found in rocks, at a great height above the sea, has been attributed to several causes. [Part 2.] By some it has been ascribed to a plastic virtue in the soil; by some, to fermentation; by some, to the influence of the celestial bodies; by some, to the casual passage of pilgrims with their scallops; by some, to birds feeding on shellfish; and by all modern geologists, with one consent, to the life and death of real mollusca at the bottom of the sea, and a subsequent alteration of the relative level of the land and sea. [Part 3.] Of these, the plastic virtue and celestial influence belong to the class of figments of fancy. [Part 4.] Causal transport by pilgrims is a real cause and might count for a few shells here and there dropped on frequented passes but is not extensive enough for the purpose of explanation. [Part 5.] Fermentation, generally, is a real cause, so far as that there is such a thing; but it is not a real cause of the production of a shell in a rock, since no such thing was ever witnessed as one of its effects, and rocks and stones do not ferment. On the other hand, for a shellfish dying at the bottom of the sea to leave its shell in the mud, where it becomes silted over and imbedded, happens daily; and the elevation of the bottom of the sea to become dry land has really been witnessed so often, and on such a scale, as to qualify it for a vera causa available in sound philosophy. Part 1 opens the discussion by identifying the phenomenon as an explanandum, i.e. a well-known fact that has been considered for long puzzling and for whose understanding a variety of explanantia have been considered. Toulmin’s ideas as a philosopher of science become relevant here. What we seek to explain is always delimited by our ideals of natural order [22] that define what is and is not puzzling. Part 2 includes the list of available explanations: H1 = soil’s plastic virtue; H2 = fermentation of soil; H3 = celestial influence; H4 = pilgrims’ passage and dropping; H5 = birds transport; H6 = life and death of real mollusca plus sea bottom alteration. Part 2 also offers the first comparative weighing of them based on authority. Note that the discussion could have easily ended here if it took place among non-scientists. More significantly, such a selection of the best available explanation (such IBE) in terms of the comparison of the authorities supporting each of them could never be understood as a refinement (nor as the process of assessment) of a simple abduction because the initial arguments adduced in their favor and subsequently weighed would not have been abductive in the first place. This is clear in the following diagram: Part 3 expresses a counter-argument against both H1 and H3. For the sake of the analysis, we may consider that both hypotheses have been offered and supported on the basis of their explanatory power, i.e. through their presentation as conclusions of abductive arguments; thus, Then, what we have here is a counter-argument in the form of a rebuttal of their associated explanations, i.e. an attack to the ‘explanatory principles’ warranting them. The counter-argument is not very developed but characterizes such principles as ‘figments of fancy’ and so unacceptable in scientific explanations.12 Part 4 contains an attack against H4 that also applies to H5. However, this time the problem is not the unacceptable character of the explanatory principles that indirectly support them, explicitly designated as ‘real causes’ or verae causae. It is rather the random, unsystematic, character of the mechanisms invoked that makes of them unsuitable for the explanation of persistent and extensive phenomena. Such mechanisms could explain the apparition of ‘some shells here and there in high rocks in frequented passes’ but not the massive and widespread presence of shells in high places. We have, thus, a plausible, even scientifically acceptable, kind of explanation that cannot warrant a proposed abductive argument. A better definition or delimitation of the data/explanandum implies, in this case, the rebuttal and dismissal of a plausible associated explanation as insufficient for it. Finally, Part 5 contains the comparative and reasoned weighing—i.e. a reason is offered for a comparative judgment whose warrant might be easily reconstructed—of the abductive arguments supporting H2 and H6. This kind of evaluative comparison of two abductive arguments is equivalent to the weighing of their associated explanations; thus, As we have seen, not all advanced hypotheses are weighed against each other in this manner, for this is an argumentatively (and cognitively) costly procedure. Some are rejected by means of direct counter-arguments that only affect them. Argumentative weighing is reserved for those that initially pass muster. In this case, just two of the available ones, although it could be much more complicated. 7. Conclusion This is only a rather simple example and it has nevertheless showed a certain argumentative complexity, requiring from us the use of a complete toolbox of flexible notions for argument analysis. The moral we want to draw is that the argumentatively interesting part of IBE-type discussions—i.e. the way typical clauses as ‘H1 explains data better that the other available hypotheses’ or ‘H1 complies with criteria S1 … Sn’ are made acceptable—cannot be abstractly and universally sequenced. The comparative selection and defense of the best available explanation in a particular epistemic context and within a particular justificatory practice is a dialectical process, not necessarily in the sense that it involves two or more parties but in the sense that it involves the opposition of arguments and the comparative weighing of their relative strength (cf. [8]). Even in a case where this process is summarized and presented by a single author (as in the example above), it is representing and accounting for a long diachronic and usually collective discussion in which very different kinds of reasons and very different combinations of arguments and argument balance may appear. In order to understand such varied processes, the best strategy is not, I claim, to keep adding summary premises to a basic abductive scheme until it is deemed sufficiently invulnerable, but, on the contrary, to account for the way counter-argumentative structures and comparative assessment work, making explicit the inter-subjective basis of justificatory practices. Footnotes 1 Warrants may have different degrees of generality but should, in any case, be of a more general scope than the data or grounds and the conclusion of the argument, at least, according to the most extended understanding of Toulmin’s model. In any case, for an alternative interpretation, see [2]. 2 Cf. https://uam.academia.edu/hubertmarraud 3 Because of the essential connection between this kind of recognized and inferred consequences and the agents that actually perform the activities of recognizing and inferring, the relations on which they are based are called by Woods ‘epistemic relations’. These ‘epistemic relations’, going beyond formal derivation, are akin to the kinds of substantive relations expressed by Toulminian warrants. 4 The social, interactive and communicative basis of the capacity for a ‘recognition of consequences’ (or, what is the same, a ‘recognition of reasons’) is assumed by Tindale’s [20] ‘philosophy of argument’. For Tindale, not only is arguing at the basis of reasoning (as suggested by Mercier and Sperber’s genealogical account) but also ‘being in audience’ is somehow previous and more basic than ‘operating as arguers’: ‘It is clear, as was previously indicated, that we also express our argumentative nature by asserting ourselves, that we operate as arguers. But the suggestion here is that we can operate in this way only because we first operate as audiences, because we fully live the condition of being an audience. On such terms the audience is the more fundamental argumentative experience. It also suggests that our appreciation of being arguers and entering the practice of arguing starts from a prior appreciation of being in audience [20, p. 21]. Such basic condition of ‘being in audience’ would determine the capacity ‘common to all’ (as Tindale [20, 40ss] remarks quoting Aristotle’s Rhetoric 1354a4–6) to recognize, appreciate and assess reasons (cf. [15, pp. 153–155]. 5 According to Tindale’s reading of Perelman and Olbrechts-Tyteca’s proposals [20, pp. 58–78], audiences and auditors may recognize and assess argumentative validity (while experiencing argumentative effectiveness) because they belong within a community that argues, that has actually been arguing long enough and that continuously revisits its standards of reasonableness. The argument offered in one particular occasion is, therefore, based on precedent (and it is interpreted and assessed as such) and aspires, in its turn, to become precedent itself (cf. [15, pp. 162–163]. 6 Searle 2001, (Chap. 3) characterizes rationality as the conscious exercise of giving and conceding reasons that precisely takes place in the presence of gaps and the absence of algorithms or necessary links between contents. The defeasible and non-monotonic character of the relations of consequence exploited in argument (i.e. their fragility, in Woods’ terms), the possibility of keeping asking for reasons that may reinforce or clarify proposed links or even show alternative ways of supporting discussed claims, the lack of compulsion to accept or enact argumentatively supported conclusions (i.e. akrasia) constitute the kinds of gaps that, for Searle, are rather possibility conditions of rationality instead of philosophical problems for its explication. 7 Woods’ [27] claim that we cannot talk about a complete ‘preservation of ignorance’ in the case of abduction because, even assuming that it does not provide evidence for believing its conclusion (the hypothesis) it does provide reasons for its selection for testing (thus changing its epistemic status) is, I think, rather in line with this idea of minimal justification, possibly understood as a more pragmatic that theoretical justification, as suggested in Section 2. 8 A corollary of my approach is that IBE is neither a characteristic type of inference nor a stereotyped argumentative sequence, so the label IBE should be rejected. 9 See [19, 9]. 10 See [14]. 11 A rather more complete analysis of this same example has already been published in Spanish [16]. 12 Contemporary philosophical discussions on the notion of ‘mechanism’ as what may be the basis of a scientific explanation [4, 5] are the updated version of Herschel’s discussion of what constitutes a vera causa. These discussions look for criteria for (and so reasons to support) scientifically acceptable explanatory principles (these would constitute an explanation’s backing in Toulmin’s terms). References 1 M. Álvarez Kinds of Reason. An Essay in the Philosophy of Action . Oxford University Press , Oxford , 2010 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 2 L. Bermejo-Luque Toulmin’s model of argument and the question of relativism. In Arguing on the Toulmin Model , D. Hitchcock and B. Verheij , eds, pp. 71 – 85 . Springer , Dordrecht , 2006 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 3 I. Douven Abduction. In The Stanford Encyclopedia of Philosophy (Summer 2017 Edition) , E. N. Zalta , ed, Metaphysics Research Lab. Center for the Study of Language and Information. Stanford University Stanford , CA , 2017 https://plato.stanford.edu/archives/sum2017/entries/abduction/ . 4 S. Glennan and P. Illari , eds. The Routledge Handbook of Mechanisms and Mechanical Philosophy . Routledge , Oxford , 2016 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 5 S. Ioannidis and S. Psillos Defense of methodological mechanism: the case of apoptosis . Axiomathes , 27 , 601 – 619 , 2017 . Google Scholar Crossref Search ADS WorldCat 6 V. Iranzo Abduction and inference to the best explanation . Theoria , 60 , 339 – 346 , 2007 . OpenURL Placeholder Text WorldCat 7 L. Magnani Abduction, Reason, and Science . Kluwer Academic/Plenum Publishers, Processes of Discovery and Explanation , Nueva York , 2001 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 8 H. Marraud Do arguers dream of logical standards? Arguers’ dialectic vs. arguments’ dialectic . Revista Iberoamericana de Argumentación , 10 , 1 – 18 , 2010 . OpenURL Placeholder Text WorldCat 9 H. Marraud Diagramas Y Estructuras Argumentativas . Published on-line , 2016 https://www.academia.edu/ . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 10 H. Marraud Guia de Campo de Esquemas Argumentativos . Published on-line , 2017 https://www.academia.edu/ . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 11 H. Marraud De las siete maneras de contraargumentar . Quadripartita Ratio , 4 , 52 – 57 , 2017 . OpenURL Placeholder Text WorldCat 12 H. Mercier and D. Sperber Why do humans reason? Arguments for an argumentative theory . Behavioral and Brain Sciences , 34 , 57 – 111 , 2011 . Google Scholar Crossref Search ADS PubMed WorldCat 13 H. Mercier and D. Sperber The Enigma of Reason . Harvard University Press , Cambridge, MA , 2017 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 14 P. Olmos Meta-Argumentation in Deliberative Discourse: Rhetoric 1360b05-1365b21. OSSA Conference Archive. 73 , 2016 https://scholar.uwindsor.ca/ossaarchive/OSSA11/papersandcommentaries/73 . 15 P. Olmos The social nature of argumentative practices: a critical review of C. Tindale’s The Philosophy of Argument and Audience Reception . Informal Logic , 38 , 151 – 183 , 2018 . Google Scholar Crossref Search ADS WorldCat 16 P. Olmos Un enfoque argumentativo sobre la abducción y sobre la ponderación de hipótesis explicativas . Theoria , 34 , 5 – 31 , 2019 . Google Scholar Crossref Search ADS WorldCat 17 F. Paglieri Review of D. Walton, Abductive Reasoning . Informal Logic , 24 , 271 – 277 , 2004 . Google Scholar Crossref Search ADS WorldCat 18 R. Pinto Argument, Inference and Dialectic. Collected Papers on Informal Logic . Kluwer Academic Publishers , Dordrecht , 2001 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 19 F. Snoeck Henkemans State-of-the-art: the structure of argumentation . Argumentation , 14 , 447 – 473 , 2000 . Google Scholar Crossref Search ADS WorldCat 20 C. Tindale The Philosophy of Argument and Audience Reception . CUP , Cambridge , 2015 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 21 S. E. Toulmin The Uses of Argument . CUP , Cambridge , 1958 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 22 S. E. Toulmin Foresight and Understanding: An Enquiry into the Aims of Science . Indiana University Press , Bloomington , 1961 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 23 J. H. M. Wagemans Argumentative patterns for justifying scientific explanations . Argumentation , 30 , 97 – 108 , 2016 . Google Scholar Crossref Search ADS WorldCat 24 D. N. Walton Abductive Reasoning . The University of Alabama Press , Tuscaloosa, AL , 2004 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 25 D. N. Walton and E. C. W. Krabbe Commitment in Dialogue: Basic Concepts of Interpersonal Reasoning . State University of New York Press , Albany , 1995 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 26 J. Woods The fragility of argument. In The Psychology of Argument. Cognitive Approaches to Argumentation and Persuasion , F. Paglieri , L. Bonelli and S. Felletti , eds, pp. 99 – 128 . College Publications (Studies in Logic 59) , London , 2016 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 27 J. Woods Reorienting the logic of abduction. In Springer Handbook of Model-Based Science , L. Magnaniy and T. Bertolotti , eds. Springer , Dordrecht , 2016 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 28 S. Yu and F. Zenker Peirce Knew why Abduction Isn’t IBE. A Scheme and Critical Questions for Abductive Argument, Argumentation , 2017 published on-line November 2017: https://doi.org/10.1007/s10503-017-9443-9 . 29 Searle, John Rationality in Action , Cambridge, MS : The MIT Press . 2001 . 30 H. Marraud Es. lógic@ Análisis y evaluación de argumentos. Madrid, Cátedra , 2013 . 31 F. John and W. Herschel Preliminary Discourse on the Study of Natural Philosophy, London: Longman, Brown, Green and Longmans , 1851 https://archive.org/details/preliminarydisco00hersiala/page/n1. © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permission@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Abduction and comparative weighing of explanatory hypotheses: an argumentative approach
JF - Logic Journal of the IGPL
DO - 10.1093/jigpal/jzz038
DA - 2020-06-01
UR - https://www.deepdyve.com/lp/oxford-university-press/abduction-and-comparative-weighing-of-explanatory-hypotheses-an-sCD0tsTLhh
DP - DeepDyve
ER -