Analog Representation and the Parts Principle

Abstract

Analog representation is often cast in terms of an engineering distinction between smooth and discrete systems. The engineering notion cuts across interesting representational categories, however, so it is poorly suited to thinking about kinds of representation. This paper suggests that analog representations support a pattern of interaction, specifically open-ended searches for content across levels of abstraction. They support the pattern by sharing a structure with what they represent. Continuous systems that satisfy the engineering notion are exemplars of this kind because they are uninterpretable unless they are structure-preserving. Analog representations, so understood, include pictures, images, diagrams, and most graphs. This conception of analogicity also fits well with a line of thought about what makes perceptual states distinctive: they satisfy a “parts principle”.

This is a preview of subscription content, access via your institution.

Notes

  1. 1.

    Haugeland’s paper “Analog and analog” appeared in 1981. Page references are to the anthologized version (Haugeland 1998).

  2. 2.

    Perhaps some inscriptions are such that we cannot tell whether they are ‘C’s, ‘G’s, or neither. In that case, the Roman alphabet would fail the test of syntactic finite differentiation. See Goodman (1968, 137–139) for discussion. This is not directly relevant to syntactic density, however, so we can leave the point here.

  3. 3.

    Matthew Katz (2008, 405) also stresses the fact, following Goodman, that in analog representation we cannot know the syntactic identity of any given representation, but he, like Goodman, didn’t notice any semantic consequences of this fact. As we will see in the next section, this is precisely where we find the action in understanding analog representation.

  4. 4.

    Haugeland conflated the second and third points just made. Analog systems are sensitive, in Haugeland’s sense, when “not only are all variations allowed, but they all matter.” (1998, 83). In line with the second point above, they all matter syntactically, and they also, in line with the third point, matter semantically. At the end of this section it will be clear that this conflation is reasonable, but to see why it’s reasonable these two conditions need to be kept apart.

  5. 5.

    This suggestion comports poorly with the mechanism behind a mercury thermometer, but we can leave that point to one side for present purposes.

  6. 6.

    Bach offers this specifically as an amendment to Goodman’s account of pictorial representation. Goodman thought that pictorial representation was simply a species of syntactically and semantically dense representation, in which relatively many qualities—color, shape, shading—matter to a representation’s identity. Pictures are relatively replete (Goodman 1968, 230).

  7. 7.

    Haugeland avoids semantics talk in his article, because “in my view digital [and analog] devices are not necessarily representational or symbolic” (1998, 88n2). The position of a potentiometer does not represent the brightness of the lights, but dimmers are analog devices. Notice that even if some analog devices are not representational, analogicity is essentially relational. Changes in the states of one thing relate to changes in the states of another in a systematic, structure-preserving fashion. This paper focuses on analog representation, but the phenomenon is not limited to representation.

  8. 8.

    As one might expect, these two are related, which we will see. Coming at things from a different angle, Philip Kitcher and Achille Varzi say pictures and many maps are worth “a vast infinity of sentences.” (2000, 377)

  9. 9.

    Notice that there are not indefinitely many abstractions available with the thermometer that has one-degree resolution. This is because there are not indefinitely many ways of abstracting over the finitely many readings the thermometer can muster. There are more than we could ever care about, but there are not indefinitely many. This is how resolution—the maximum information delivered by any given reading—interacts with the number of available abstractions.

  10. 10.

    I am not claiming that a representation is digital if it fails to be analog. No account of digital representation is on offer here. Analog and digital make a nice contrast, but no account of the distinction suggests that they jointly cover the range of representation. So, one needn’t offer an account of both at once.

  11. 11.

    These claims are all formulated in terms of “pictures”, but they can actually be understood a lot more broadly, as we will see. Casati and Varzi (1999) appeal to a related claim in working out a semantics of maps and Kulvicki (2006, 59n2) discusses it in relation to pictorial transparency. See Balog (2009) for a critical discussion of Fodor’s proposal and Kulvicki (2014, Ch 8) where I discuss the following way to think of the parts principle, though I did not notice the connection there to analog representation.

  12. 12.

    Fred Dretske (1981, 135–141) appropriates the analog-digital distinction in the service of understanding the distinction between sensation, on the one hand, and thought-like states, on the other. He uses the terms ‘analog’ and ‘digital’ in a distinctive manner that does not fit with the present proposal. Elsewhere, however, I suggest a way of emending his view that makes it easy to show how the present conception of analog representation fits well with his account of sensory states (Kulvicki 2004, 383–389; 2007, 361–364).

  13. 13.

    I thank two anonymous referees, the editors of this special issue, and Eliot Michaelson for their very useful comments. Audiences at Dartmouth College, Rice University, the University of Auckland, and the Kunstakademie Düsseldorf also provided helpful feedback. This research was supported by a Dartmouth College Scholarly Research and Advancement Award.

References

  1. Alberti, L. 1991 (1435). On painting. C. Grayson, trans. London: Penguin.

  2. Bach, K. 1970. Part of what a picture is. British Journal of Aesthetics 10: 119–137.

    Article  Google Scholar 

  3. Balog, K. 2009. Jerry Fodor on non-conceptual content. Synthese 170: 311–320.

    Article  Google Scholar 

  4. Blachowicz, J. 1997. Analog representation beyond mental imagery. Journal of Philosophy 94(2): 55–84.

    Article  Google Scholar 

  5. Casati, R., and A. Varzi. 1999. Parts and places. Cambridge: MIT Press.

    Google Scholar 

  6. Dretske, F. 1981. Knowledge and the flow of information. Cambridge: MIT Press.

    Google Scholar 

  7. Fodor, J. 2008. LOT2: The language of thought revisited. Oxford: Clarendon.

    Google Scholar 

  8. Goodman, N. 1968. Languages of art. Indianapolis: Hackett.

    Google Scholar 

  9. Goodman, N. 1972. Seven strictures on similarity. In Problems and projects, 437–446. New York: Bobbs-Merrel.

  10. Goodman, N., and C.Z. Elgin. 1988. Reconceptions in philosophy and other arts and sciences. London: Routledge.

    Google Scholar 

  11. Haugeland, J. 1981. Analog and analog. Philosophical Topics 12: 213–225. Anthologized in Haugeland 1998.

    Article  Google Scholar 

  12. Haugeland, J. 1991. Representational genera. In Philosophy and connectionist theory, eds. W. Ramsey, S. Stich and D. Rumelhart. Hillsdale, NJ: Lawrence Erlbaum. Reprinted in Haugeland 1998.

  13. Haugeland, J. 1998. Having thought. Cambridge: Harvard University Press.

    Google Scholar 

  14. Katz, M. 2008. Analog and digital representation. Minds & Machines 18: 403–408.

    Article  Google Scholar 

  15. Kitcher, P., and A. Varzi. 2000. Some pictures are worth 2-to-the-aleph-null sentences. Philosophy 75: 377–381.

    Article  Google Scholar 

  16. Kosslyn, S. 1980. Image and mind. Cambridge: Harvard University Press.

    Google Scholar 

  17. Kosslyn, S. 1994. Image and brain. Cambridge: MIT Press.

    Google Scholar 

  18. Kulvicki, J. 2004. Isomorphism in information-carrying systems. Pacific Philosophical Quarterly 85: 380–395.

    Article  Google Scholar 

  19. Kulvicki, J. 2006. On images: Their structure and content. Oxford: Clarendon.

    Google Scholar 

  20. Kulvicki, J. 2007. Perceptual content is vertically articulate. American Philosophical Quarterly 44(4): 357–369.

    Google Scholar 

  21. Kulvicki, J. 2010. Knowing with images: medium and message. Philosophy of Science 77(2): 295–313.

    Article  Google Scholar 

  22. Kulvicki, J. 2014. Images. London: Routledge.

    Google Scholar 

  23. Lewis, D. 1971. Analog and digital. Noûs 5(3): 321–327.

    Article  Google Scholar 

  24. Maley, C. 2011. Analog and digital, continuous and discrete. Philosophical Studies 155: 117–131.

  25. Moor, J. 1978. Three myths of computer science. British Journal of the Philosophy of Science 29: 213–222.

    Article  Google Scholar 

  26. Shepard, R. 1978. The mental image. American Psychologist 33: 125–137.

  27. Sober, E. 1976. Mental representations. Synthese 33(1): 101–148.

    Article  Google Scholar 

  28. Tye, M. 1991. The imagery debate. Cambridge: MIT Press.

    Google Scholar 

  29. Van Essen, D., W. Newsome, and J. Maunsell. 1984. The visual field representation in striate cortex of the macaque monkey. Vision Research 24: 429–448.

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to John Kulvicki.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kulvicki, J. Analog Representation and the Parts Principle. Rev.Phil.Psych. 6, 165–180 (2015). https://doi.org/10.1007/s13164-014-0218-z

Download citation

Keywords

  • Perceptual State
  • Representational Content
  • Color Photo
  • Fourth Condition
  • Analog Representation