Abstract
In this paper I apply the mechanistic account of explanation to engineering science. I discuss two ways in which this extension offers further development of the mechanistic view. First, functional individuation of mechanisms in engineering science proceeds by means of two distinct sub types of role function, behavior function and effect function, rather than role function simpliciter. Second, it offers refined assessment of the explanatory power of mechanistic explanations. It is argued that in the context of malfunction explanations of technical systems, two key desiderata for mechanistic explanations, ‘completeness and specificity’ and ‘abstraction’, pull in opposite directions. I elaborate a novel explanatory desideratum to accommodate this explanatory context, dubbed ‘local specificity and global abstraction’, and further argue that it also holds for mechanistic explanations of malfunctions in the biological domain. The overall result is empirically-informed understanding of mechanistic explanation in engineering science, thus contributing to the ongoing project of understanding mechanistic explanation in novel or relatively unexplored domains. I illustrate these claims in terms of reverse engineering and malfunction explanations in engineering science.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
I take engineering science to be a ‘science of making’. The view of engineering as a ‘science’ can be defended in terms of key objectives that it has in common with other ‘traditional’ sciences; explanation, prediction, and understanding of complex systems, in casu technical systems. Explanation, for instance, is a key element in a variety of engineering designs methods, such as diagnostic reasoning methods, reverse engineering and redesign methods, and in knowledge base-assisted designing (see sections 3 and 4).
Several concepts of function are on offer in the philosophical literature, with selected effect and role concepts of function being the prominent ones (see Walsh and Ariew 1996). The selected effect concept of function is invoked to explain the presence or characteristics of an item (function bearer) in terms of whatever it does (selected effect) that contributed to the fitness of possessors of these items in the evolutionary past (Millikan 1989). The role concept of function is detached from evolutionary selective history. It, rather, is used to explain how an item contributes to some overall capacity of a system of which the item is a part (Cummins 1975; Walsh and Ariew 1996). The role concept of function is considered far more relevant than the selected effect concept of function for explaining how mechanisms produce explananda phenomena (Craver 2001, 2007; McKay Illari and Williamson 2010): selective histories do not explain how items are situated in mechanisms and how they contribute to the mechanisms of which they are a part (Craver 2007; McKay Illari and Williamson 2010). The role concept of function is invoked specifically for this explanatory job.
Function ascription is also considered vital for the hierarchical nesting of explanations: lower-level explanations are nested in/related to higher-level ones if the phenomena explained by lower-level explanations—say, the heart’s “pumping blood through the circulatory system”—have been identified as having a function in mechanisms described by higher-level explanations—say, the circulatory system’s mechanism for “distributing oxygen and nutrients to parts of the body” (Craver 2001).
Their analysis focused in particular on protein synthesis and natural selection.
The term ‘archetypical’ here refers to ‘most common’; the three conceptualizations of function are not meant to be exhaustive. For instance, some engineers use ‘function’ to refer to intentional behaviors of agents (see van Eck 2010), and others have recently explored the idea that ‘function’ might be a Wittgensteinian family resemblance concept (Carrara et al. 2011). In reverse engineering analyses, ‘function’ refers to actual or expected behavior, without the normative connotation ‘desired’.
Behavior and effect functions thus have a partly common semantic structure: certain aspects or features of behaviors that they both refer to. They are dissimilar in the sense that behavior function descriptions refer to additional behavioral aspects, not referred to in effect function descriptions, so as to make these descriptions accord with physical conservation laws. The relation between behavior and effect function is asymmetrical in the sense that effects, being subsets of behaviors, are straight forwardly derivable from behaviors, but not vice versa. From a given effect one cannot automatically derive the behavior of which the effect is a part. Cars that run on gas operate by means of different energy conversions than cars than run on electricity, yet both display the same effects, say, delivering acceleration. The semantic structure that they partly have in common creates the possibility and need to be pluralist about mechanistic role functions, i.e., different ways to conceive of the role functions of mechanisms, in the context of engineering science. I defend this pluralism about mechanistic role functions in section 3.2. To be sure, I am thus not advocating a pluralist view about functions of mechanisms with a completely dissimilar semantic structure. I thank an anonymous referee for pressing me on these points.
In design methodologies that advance effect and/or purpose functions the concept of behavior is typically introduced as well, alongside function. By invoking behavior descriptions of technical artifacts alongside functional descriptions, physical conservation laws are taken into account.
Single functional decomposition models in which different concepts of function are described are rare in engineering.
This is not to say that the explanandum in reverse engineering explanation, or mechanistic explanation in general, cannot be described in contrastive fashion. For instance, the request for explanation may concern why an artifact x exhibits behavior b with value y rather than behavior b with value z. However, this is a different contrast than the one drawn in the explanandum of why artifact x does not exhibit behavior b rather than displaying behavior b.
An anonymous referee rightly pointed out that the ‘omission of specifics’, as in the case of malfunction explanation, highlights a general principle of relevance that holds for most contexts in which one aims to explain an isolated feature of a system’s behavior. However, such a general principle of relevance can be interpreted in different ways depending on the desiderata for mechanistic models one adopts. In section 4 I argue that two recently proposed desiderata for mechanistic models pull in opposite directions in the context of malfunction explanation, signaling the need for a novel one. I elaborate such a desideratum in this paper.
This is the most straightforward scenario. If there are backup systems that are intended to prevent malfunction, and modern technology is replete with them, failing backup systems should be referred to as well, of course, in explaining system level malfunctions.
That is, structural and behavioral characteristics are considered irrelevant in a first round functional analysis of malfunction. After this analysis, more detailed behavioral models of components and their behaviors are used for identifying specific explanatorily relevant structural and behavioral characteristics of malfunctioning components/sub mechanisms (Bell et al. 2007). However, immediately specifying these details in functional models is taken to result in listing a lot of irrelevant details (see the FIL methodology described in this section).
Note that behavior and effect descriptions of function describe, in different ways, the contributions of components to mechanisms of which they are a part. The distinction between behavior and effect function thus is not to be conflated with the distinction between a mechanism description and a description of a mechanisms’ overall activity. Neither is the behavior-effect function distinction to be conflated with the distinction between ‘isolated’ and ‘contextual’ descriptions of an entity’s activity (Craver 2001): isolated descriptions describe activities without taking into account the mechanisms in which they are situated; contextual descriptions describe activities in terms of the mechanistic contexts in which they are situated and to which they contribute. Both behavior and effect functions are of the contextual variety, describing contributions of components to the mechanisms of which they are a part.
Although my preferred conception of mechanistic explanation is an epistemic one, both advocates of epistemic (e.g., Wright and Bechtel 2007; Wright 2012) and ontic conceptualizations (e.g., Salmon 1984; Glennan 2005; Craver 2007, 2012b) of mechanistic explanation can sign up to this project. On an epistemic reading, explanations are explanatory texts that procure understanding, whereas on an ontic reading mechanisms in the world are explanations. Illari (2013) recently argued that the ontic-epistemic dispute currently going on in the literature is (or should be understood as) moving away from analysis of the term ‘explanation’ per se, to elaborating ontic and epistemic constraints on good explanations. Good mechanistic explanations should describe (ontic) mechanisms in the world in such a fashion that (epistemic) understanding of their workings is procured. If one subscribes to this view, as I do, then acknowledging the behavior-effect function distinction is relevant to capture good mechanistic explanations in engineering science. Depending upon explanatory context, (ontic) mechanisms are individuated in different ways using different conceptualizations of function in engineering science. These different functional individuations, in terms of different function conceptualizations, are tailor-made or ‘engineered’ to the task at hand, so as to highlight the (epistemic) relevant features of technical systems-mechanisms. So whether one takes features of mechanisms, behaviors and effects, as providing (ontic) explanations or descriptions of these features as doing the (epistemic) explanatory work, acknowledging the behavior-effect distinction is relevant to both camps for understanding good explanations in engineering science. Moreover, functional individuation of mechanisms is also explicitly endorsed by authors who defend an ontic view on explanation, at least by Machamer et al. (2000). They write: “mechanisms are identified and individuated by the activities, and entities that constitute them, by their start and finish conditions, and by their functional roles. Functions are the roles played by entities and activities in a mechanism. To see an activity as a function is to see it as a component in some mechanism, that is, to see it in a context that is taken to be important, vital, or otherwise significant.” (Machamer et al. 2000, 6) So, function ascription is not only an epistemic factor affecting how we describe mechanisms: it is also an individuation condition of mechanisms (see McKay Illari and Williamson 2010).
Constitutive relevance is explicated in terms of an adapted version of Woodward’s theory, since Woodward advances his manipulability theory as an account of causal explanation, while constitutive relevance is explicitly defined as a non-causal relationship (see Craver 2007, 153–154; Craver and Bechtel 2007, 552–554).
To be sure, as the quotation makes clear, Levy and Bechtel acknowledge that in other explanatory contexts structural details can be important.
This is Levy and Bechtel’s (2013) interpretation of Machamer et al. (2000). I think that their reading is correct, for Machamer et al. (2000) write: “We introduce the term “mechanism schema” for an abstract description of a type of mechanism. A mechanism schema is an abstract truncated description of a mechanism that can be filled with descriptions of known component parts and activities.” (Machamer et al. 2000, 15). They elaborate: “When instantiated, mechanism schemata yield mechanistic explanations of the phenomenon that the mechanism produces.” (p. 17) […] “Sometimes a sketch has to be abandoned in the light of new findings. In other cases it might become a schema, serving as an abstraction that can be instantiated as needed for the tasks mentioned above, e.g., explanation, prediction, and experimental design.” (p. 18) So, instantiation—filling in details of entities and activities—is what turns a schema into an explanation according to Machamer et al. (2000).
To be sure, in Strevens’ (2004, 2008) system, explananda phenomena may comprise many things, such as events, properties, and regularities. What is crucial in Strevens’ system is that explanatory models should only refer to those factors that are crucial for the explanatory target to obtain, i.e., to occur, whether it be a specific event, a regularity, a property, or something else. Strevens (2004, 158) himself puts it thus: “the explanatorily relevant parts of any causal network are the elements that made a difference to whether or not the explanandum occurred. It is important to note the whether or not. To be explanatorily relevant, a causal factor must not merely make a difference to how the explanandum occurred; it must make a difference large enough to bear on whether or not it occurred at all.” Note that explananda are not restricted to events, properties, regularities, and the like, occurring or not but, rather, that explanatorily relevant factors must make a difference large enough for explanatory targets to obtain.
Parts of this section draw on (van Eck 2014).
The notion of monitoring the effects of single component removals on overall behaviors of technical systems as is done in reverse engineering (see section 2), corresponds to the bottom-up condition of the mutual manipulability account of changing the overall behavior by intervening to change an entity’s activity (see section 3.1).
Not incidentally, the concept of function employed in the Kitamura-Mizoguchi methodology is that of effect function.
This is in keeping with engineering practice. After a first round functional analysis of malfunction, more detailed behavioral models of components and their behaviors are used in FIL for assessing specific structural characteristics of malfunctioning components (Bell et al. 2007).
I adapt this example from (Nervi 2010)
I adopt the term ‘behavioral explanation’ from Matthewson and Calcott (2011).
I am not claiming that input–output descriptions are always explanatory. Yet, the manner in which input and output are specified in FIL-functional descriptions does confer explanatory traction on such descriptions, since they provide course-grained answers to the question ‘why malfunction, rather than normal function?’.
References
Barros, B. (2009). Natural selection as a mechanism. Philosophy of Science, 75(3), 306–322.
Bechtel, W. (2006). Discovering cell mechanism; The creation of modern cell biology. Cambridge: CUP.
Bechtel, W. (2008a). Mechanisms in cognitive psychology: what are the operations? Philosophy of Science, 75, 983–994.
Bechtel, W. (2008b). Mental mechanisms: philosophical perspectives on cognitive neuroscience. London: Routledge.
Bechtel, W., & Abrahamson, A. (2005). Explanation: a mechanist alternative. Studies in History and Philosophy of Biological and Biomedical Sciences, 36, 421–441.
Bell, J., Snooke, N., & Price, C. (2007). A language for functional interpretation of model based simulation. Advanced Engineering Informatics, 21, 398–409.
Calcott, B. (2009). Lineage explanations: explaining how biological mechanisms change. British Journal for the Philosophy of Science, 60, 51–78.
Carrara, M., Garbacz, P., & Vermaas, P. E. (2011). If engineering function is a family resemblance concept: assessing three formalization strategies. Applied Ontology, 6, 141–163.
Chakrabarti, A., & Bligh, T. P. (2001). A scheme for functional reasoning in conceptual design. Design Studies, 22, 493–517.
Chandrasekaran, B., & Josephson, J. R. (2000). Function in device representation. Engineering with Computers, 16, 162–177.
Craver, C. F. (2001). Role functions, Mechanisms, and Hierarchy. Philosophy of Science, 68, 53–74.
Craver, C. F. (2007). Explaining the brain: mechanisms and the mosaic unity of neuroscience. New York: Oxford University Press.
Craver, C. F. (2012a). Functions and mechanisms: A perspectivalist account. In P. Huneman (Ed.), Functions (pp. 133–158). Dordrecht: Springer.
Craver, C. F. (2012b). Explanation: The ontic conception. In A. Hutteman & M. Kaiser (Eds.), Explanation in the biological and historical sciences. Berlin: Springer.
Craver, C. F., & Bechtel, W. (2005). Mechanisms and mechanistic explanation. In S. Sarkar & J. Pfeiffer (Eds.), The philosophy of science: an encyclopedia (pp. 469–478). New York: Routledge.
Craver, C. F., & Bechtel, W. (2007). Top-down causation without top-down causes. Biology and Philosophy, 22, 547–563.
Craver, C. F., & Darden, L. (2001). Discovering mechanisms in neurobiology. The case of spatial memory. In P. Machamer et al. (Eds.), Theory and method in the neurosciences. Pittsburgh: University of Pittsburg Press.
Cummins, R. (1975). Functional analysis. Journal of Philosophy, 72, 741–765.
Darden, L. (2006). Reasoning in biological discoveries. Cambridge: Cambridge University Press.
Darden, L., & Craver, C. F. (2002). Strategies in the interfield discovery of the mechanism of protein synthesis. Studies in the History and Philosophy of the Biological and Biomedical Sciences, 33, 1–28.
De Ridder, J. (2006). Mechanistic artefact explanation. Studies in History and Philosophy of Science, 37, 81–96.
Deng, Y. M. (2002). Function and behavior representation in conceptual mechanical design. Artificial Intelligence for Engineering Design, Analysis, and Manufacturing, 16, 343–362.
Erden, M. S., Komoto, H., Van Beek, T. J., D’Amelio, V., Echavarria, E., & Tomiyama, T. (2008). A review of function modeling: approaches and applications. Artificial Intelligence for Engineering Design, Analysis, and Manufacturing, 22, 147–169.
Gervais, R., & Weber, E. (2013). Plausibility versus richness in mechanistic models. Philosophical Psychology, 26(1), 139–152.
Glennan, S. (2005). Modeling mechanisms. Studies in the History and Philosophy of the Biological and Biomedical Sciences, 36(2), 375–388.
Glennan, S. (2010). Ephemeral mechanisms and historical explanation. Erkenntniss, 72, 251–266.
Hawkins, P. G., & Woollons, D. J. (1998). Failure modes and effects analysis of complex engineering systems using functional models. Artificial Intelligence in Engineering, 12(4), 375–397.
Illari, P. (2013). Mechanistic explanation: integrating the ontic and epistemic. Erkenntnis, online first. doi:10.1007/s10670-013-9511-y.
Kaplan, D. M., & Bechtel, W. (2011). Dynamical models: an alternative or complement to mechanistic explanations? Topics in Cognitive Science, 3, 438–444.
Kaplan, D., & Craver, C. (2011). The explanatory force of dynamical and mathematical models in neuroscience: a mechanistic perspective. Philosophy of Science, 78, 601–627.
Kitamura, Y., Koji, Y., & Mizoguchi, R. (2005). An ontological model of device function: industrial deployment and lessons learned. Applied Ontology, 1, 237–262.
Levy, A., & Bechtel, W. (2013). Abstraction and the organization of mechanisms. Philosophy of Science, 80, 241–261.
Lipton, P. (1993). Making a difference. Philosophica, 1, 39–54.
Machamer, P. K., Darden, L., & Craver, C. F. (2000). Thinking about mechanisms. Philosophy of Science, 57, 1–25.
Matthewson, J., & Calcott, B. (2011). Mechanistic models of population-level phenomena. Biology and Philosophy, 26(5), 737–756.
McKay Illari, P., & Williamson, J. (2010). Function and organization: comparing the mechanisms of protein synthesis and natural selection. Studies in History and Philosophy of Biological and Biomedical Sciences, 41, 279–291.
McKay Illari, P., & Williamson, J. (2012). What is a mechanism? Thinking about mechanisms across the sciences. European Journal for Philosophy of Science, 2, 119–135.
Millikan, R. (1989). In defense of proper functions. Philosophy of Science, 56, 288–302.
Moghaddam-Taaheri, S. (2011). Understanding pathology in the context of physiological mechanisms: the practicality of a broken-normal view. Biology and Philosophy, 26, 603–611.
Nervi, M. (2010). Mechanism, malfunctions and explanation in medicine. Biology and Philosophy, 25, 215–228.
Otto, K. N., & Wood, K. L. (1998). Product evolution: a reverse engineering and redesign methodology. Research in Engineering Design, 10, 226–243.
Otto, K. N., & Wood, K. L. (2001). Product design: techniques in reverse engineering and new product development. Upper Saddle River: Prentice Hall.
Piccinini, G., & Craver, C. F. (2011). Integrating psychology and neuroscience: functional analyses as mechanism sketches. Synthese, 183, 283–311.
Price, C. J. (1998). Function-directed electrical design analysis. Artificial Intelligence in Engineering, 12(4), 445–456.
Salmon, W. (1984). Scientific explanation: three basic conceptions. Proceedings of the Biennial Meeting of the Philosophy of Science Association, 2, 293–305.
Skipper, R., & Milstein, R. (2005). Thinking about evolutionary mechanisms: natural selection. Studies in History and Philosophy of Biological and Biomedical Sciences, 36, 327–347.
Stone, R. B., & Chakrabarti, A. (2005). Guest editorial. Special issue: engineering applications of representations of function, part 2. Artificial Intelligence for Engineering Design, Analysis and Manufacturing, 19(3), 137.
Stone, R. B., & Wood, K. L. (2000). Development of a functional basis for design. Journal of Mechanical Design, 122, 359–370.
Strevens, M. (2004). The causal and unification approaches to explanation unified—causally. Noûs, 38(1), 154–176.
Strevens, M. (2008). Depth: An account of scientific explanation. Cambridge, MA: Harvard University Press.
Sustar, P. (2007). Neo-functional analysis: phylogenetical restrictions on causal role functions. Philosophy of Science, 74, 601–615.
Thagard, P. (2003). Pathways to biomedical discovery. Philosophy of Science, 70, 235–254.
Van Eck, D. (2010). On the conversion of functional models: bridging differences between functional taxonomies in the modeling of user actions. Research in Engineering Design, 21(2), 99–111.
Van Eck, D. (2011). Supporting design knowledge exchange by converting models of functional decomposition. Journal of Engineering Design, 22(11–12), 839–858. doi:10.1080/09544828.2011.603692.
Van Eck, D. (2014). Validating function-based design methods: an explanationist perspective. Philosophy and Technology, online first. doi:10.1007/s13347-014-0168-5.
Van Eck, D., & Weber, E. (2014). Function ascription and explanation: elaborating an explanatory utility desideratum for ascriptions of technical functions. Erkenntnis, 79, 1367–1389. doi:10.1007/s10670-014-9605-1.
Vermaas, P. E. (2009). The flexible meaning of function in engineering. Proceedings of the 17th International Conference on Engineering Design (ICED 09), 2, 113–124.
Walsh, D. M., & Ariew, A. (1996). A taxonomy of functions. Canadian Journal of Philosophy, 26(4), 493–514.
Weisberg, M. (2007). Three kinds of idealization. The Journal of Philosophy, 104(12), 639–659.
Woodward, J. (2003). Making things happen. Oxford: Oxford University Press.
Wright, C. D. (2012). Mechanistic explanation without the ontic conception. European Journal for the Philosophy of Science, 2, 375–394.
Wright, C., & Bechtel, W. (2007). Mechanisms and psychological explanation. In P. Thagard (Ed.), Philosophy of psychology and cognitive science (pp. 31–79). Amsterdam: Elsevier.
Acknowledgments
I am grateful for helpful comments by Conor Dolan, Phyllis Illari, Bert Leuridan, and Huib Looren de Jong. Comments by two anonymous referees proved particularly rewarding.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
van Eck, D. Mechanistic explanation in engineering science. Euro Jnl Phil Sci 5, 349–375 (2015). https://doi.org/10.1007/s13194-015-0111-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13194-015-0111-3