Universities from an epistemological point of view

Daniel Stoljar

When Prime Minister of Australia Scott Morrison called in 2020 for a reassessment of the social contract between universities and the rest of society, it was easy to be sceptical: politics was just below the surface. Still, whatever the underlying intent, the issue of the nature and social function of universities is an important and urgent one to address, and it is good that Morrison put it on the agenda.¹

From the perspective of my own discipline of philosophy, what is most apparent are the deep epistemological questions underlying the issue. Universities, as the philosopher David Lewis once put it, are dedicated, at least in the ideal case, to ‘the advancement of knowledge: its transmission by teaching, its expansion by research’.² If there is a social justification for universities, it must be that the state legitimately wants what universities have, and what universities seem to have is knowledge.

The moment we try to turn this vague thought into a coherent proposal about the social function of universities, however, we run into a major problem. Universities certainly contain lots of people who expand knowledge by research and transmit it by teaching; moreover, they do this almost entirely supported by the state and under conditions of considerable freedom. You might expect, therefore, that, if universities have a social function, the knowledge in question will be usable in some obvious way—by people obtaining jobs, by corporations and industry, by governments. But, while this sometimes happens, and while it is a good thing when it does, this is not the universal or even typical case, as a quick look at any academic journal will show. Across the research university, in both the sciences and the humanities, knowledge is driven by internal disciplinary pressures, is usually extremely technical or at least very complicated, and is often incomprehensible to the uninitiated.

Why, then, do institutions of this general character play the social roles they do? What is in it for the state?

Of course, questions like this have been asked many times in the past. I will look here at one particular answer and how it has changed over time, with a view not just to its inherent plausibility but also to its underlying epistemological assumptions.

This answer was offered at the start of the nineteenth century by German idealist philosophers such as Fichte, Schelling and von Humboldt.³ They argued that if you want to understand the social function of universities, you should concentrate not on the objects of knowledge—that is, on the theories, models or claims advanced by particular individuals at particular times, nor on whether such things are usable. Rather, you should focus on the individuals who know such things—epistemic agents, we might call them—and on the distinctive characteristics and capacities of such agents. The basic idea is that the state requires epistemic agents of various sorts on a large scale—for example, to populate the civil service—and research universities are the only institutions that reliably produce them.

It was Wilhelm von Humboldt’s version of this answer that became dominant. I think his view may be stated as follows. Suppose we have an ideal and infinite agent who knows everything about the totality of existence, not just its empirical physical aspects, but its moral, social and psychological aspects too. Consider now the individuals, both students and faculty, who make up a university—individuals, it need hardly be said, who are non-ideal and non-infinite. What such individuals aim to do, in von Humboldt’s picture, is to align themselves to the greatest extent possible with this ideal agent. Research is a matter of tracing out paths to knowledge that, while new to ordinary non-ideal agents, have already been taken by their ideal counterpart; teaching is a matter of cooperatively retracing older paths. Anyone, and so any student, who immerses themselves in such processes will become the sort of epistemic agent valued by the state. Hence universities ‘join objective knowledge with the formation of the subject’, as von Humboldt put it.⁴

This picture of universities and their social role has a number of striking elements. There is no deep distinction between research and teaching; both are ways to achieve the same underlying goal. Nor is there any division between natural sciences on the one hand, and humanities and social sciences on the other. Since the knowledge of the ideal agent is unified, it makes little difference what part of it you align with. The picture also provides a basis for academic freedom: given that the state needs epistemic agents, it is in its own interest to leave universities alone to reliably produce them.

It is not hard to see why the state in the figure of King Friedrich Wilhelm III accepted this line of reasoning, which it did when the University of Berlin was founded in 1810. Aside from its intrinsic features, one suspects several further assumptions were operating in the background that made it irresistible. One is that the ideal agent present in von Humboldt’s formulation is of a particular cultural type—a Prussian gentlemen, perhaps, or a German one. Another is that the agent is not simply a standard by which knowledge is measured but constitutes the world itself; this is partly what ‘idealism’ means in the phrase ‘German idealism’. From this vantage point, the university, the state and the world are in metaphysical harmony.

Nevertheless, it would be wrong to dismiss von Humboldt’s answer just because it contains cultural and philosophical assumptions that, for us, are objectionable. A better move is to retain the basic idea but remove the baggage.

One such view was offered by American pragmatist philosopher John Dewey a hundred years after von Humboldt, in the 1915 Declaration of Principles of the American Association of University Professors, a body formed by Dewey himself.⁵ It is challenging to give a coherent statement of Dewey’s pragmatism beyond saying that it is not German idealism. For him, key notions such as truth, knowledge and inquiry are subordinate to democracy in some fundamental sense, though what that really means remains elusive.⁶

Still, Dewey’s position as regards universities is reasonably straightforward, and that is because it is structurally like von Humboldt’s. For Dewey, as for von Humboldt, universities are the only institutions that can provide the state what it needs—the difference is that, for von Humboldt, the state is an ethno-nationalist one, while for Dewey it is a democratic one.

Why does the democratic state need epistemic agents? In part, Dewey’s answer is the old one: ‘for various branches of the public service’.⁷ But he also offers a more interesting reason: universities solve the basic problem of democratic states, namely, that while such states derive their ultimate authority from the public, they must at the same time avoid what Dewey calls ‘a tyranny of public opinion’.⁸ Dewey does not assume that the American state circa 1915 is democratic, but he thinks that democratic states in their nature require the epistemic agents that only universities produce.

A different version of the Humboldtian justification appeared 30 years after Dewey’s declaration in the form of Vannevar Bush’s Science: The Endless Frontier.⁹ This was a report to the US president that Bush produced in his role as director of the Office of Scientific Research and Development immediately after World War II. Bush did not emphasise pressures internal to the state, such as the need for public servants or the threat of public opinion, though what he wrote was consistent with this. His point, rather, was that the US in 1945 confronted, and would continue to confront, huge external challenges (e.g. from global disease or hostile regimes). To respond to those challenges, the state would again require epistemic agents on a large scale, something that only universities could provide.

Unlike von Humboldt and Dewey, Bush was no philosopher; he was an MIT-trained engineer turned Cold War planner. Yet the background epistemology of his report aligned (whether by design or chance) with a then dominant one in academic philosophy, namely, positivism, particularly the logical positivism of philosophers such as Rudolf Carnap who had, a few years earlier, escaped Europe for America.¹⁰ Positivism can mean several things, but here it means that genuine knowledge or science is restricted to the sort produced in the natural sciences and mathematics; whatever happens in the humanities may be okay in its way, but it is not the real thing. This is quite different from von Humboldt, who classified history and philosophy, for example, as sciences (‘Wissenschaften’) alongside physics and chemistry.

Bush’s combination of the Humboldtian view with positivism leaves the humanities and social sciences badly exposed. Suppose universities have, as their internal aim, the expansion and transmission of knowledge and, as their external social aim, the production of epistemic agents needed by the state. If the notions of ‘knowledge’ and ‘epistemic agent’ are understood in a restricted positivist way, we lack a justification for universities insofar as they concern the humanities. Maybe this does not matter if money is plentiful, as it was in the US in the initial period after the war. But, at other times, the humanities will inevitably seem an optional extra, nice to have around but not strictly necessary.

How to react? One might think this is the end of the line for von Humboldt and that the humanities stand in need an alternative defence (e.g. as keepers of traditions of thought).¹¹ But a different reaction is, once again, to keep the Humboldtian structure but rethink the background epistemology.

For there is nothing inevitable about positivism. Within academic philosophy it is mostly seen as outdated as German idealism or Deweyite pragmatism. The idea, in particular, that you should restrict notions of knowledge or inquiry to natural sciences is widely rejected; there is no coherent notion of ‘natural’ or ‘science’ that grounds such a restriction. Moreover, an injunction to privilege natural science disciplines on practical grounds runs into the problem that many parts of such disciplines have no immediate relevance, and many areas outside such disciplines do—the relation of linguistics to large language models and artificial intelligence is a recent example.

Suppose then we try to offer anew the Humboldtian view of universities, this time paired with twenty-first-century epistemology. What would it look like?

That is an open and exciting question. The nature of universities has been widely discussed by historians and sociologists, but contemporary philosophers pay it little attention.¹² My own view is that a position of this sort would take something from each of the three incarnations of von Humboldt that we have looked at but move beyond them. From Dewey, it would take the view that universities play a necessary role in democracy. From Bush, it would take the view that universities are required as insurance against external challenges, whatever they happen to be.¹³ And, from von Humboldt himself, it would take the view that there is no philosophical justification for restricting terms like ‘science’ or ‘knowledge’ to natural science or knowledge.

That is a plan for a position, of course, rather than a position. But if we could make it out clearly, we would have an answer to Morrison’s challenge about universities.

---

1. 1 Morrison’s challenge and related issues are discussed at length in Tamson Pietsch’s podcast The New Social Contract (2020), as well as in J. Horne and M. A. M. Thomas, eds, Australian Universities: A Conversation about the Public Good (Sydney: Sydney University Press, 2022).

2. 2 David Lewis, Papers in Ethics and Social Philosophy (Cambridge: Cambridge University Press, 2000), 187.

3. 3 The relevant writings of these authors are gathered in Part I of Louis Menand, Paul Reitter and Chad Wellmon, eds, The Rise of the Research University: A Sourcebook (Chicago: University of Chicago Press, 2017). Here and throughout, I will state the ideas in my own way rather than aiming at scholarly accuracy.

4. 4 See Menand, Reitter and Wellmon, The Rise of the Research University, 108.

5. 5 Edwin Seligman, Charles E. Bennett, James Q. Dealey, Richard T. Ely, Henry W. Farnam, Frank A. Fetter, Franklin H. Giddings, Charles A. Kofoid, Arthur O. Lovejoy, Frederick W. Padelford, Roscoe Pound, Howard C. Warren and Ulysses G. Weatherly, ‘AAUP’s 1915 Declaration of Principles’, Bulletin of the American Association of University Professors 40, no. 1 (1954): 90–112, doi.org/10.2307/40220998. The declaration is usually attributed to a committee chaired by Seligman, but since Dewey was the president of the association at the time it was adopted and selected the committee in the first place, I will take the liberty in the text of speaking as if Dewey was the author.

6. 6 A recent attempt to state Dewey’s pragmatism is Philip Kitcher, Preludes to Pragmatism: Toward a Reconstruction of Philosophy (Oxford: Oxford University Press, 2012).

7. 7 See Seligman, ‘AAUP’s 1915 Declaration of Principles’, 99.

8. 8 Ibid., 103.

9. 9 Vannevar Bush, Science: The Endless Frontier (Washington: United States Printing Office, 1945). For an illuminating discussion of Bush’s report, see Richard Lewontin, ‘The Cold War and the Transformation of the Academy’, in The Cold War and the University, ed. A. Schiffrin (New York: New Press, 1997), 1–34. For some comparable ideas in the Australian context, see Michael Goodman, ‘Let Us Have More Scientists, and More Humanists, in Australian Universities: A Conversation about the Public Good, ed., J. Horne and M. A. M. Thomas (Sydney: Sydney University Press, 2022).

10. 10 See, for example, Rudolf Carnap, The Logical Syntax of Language (London: Kegan Paul, 1937).

11. 11 A recent defence of this sort of position is Eric Adler, The Battle of the Classics: How a Nineteenth-Century Debate Can Save the Humanities Today (New York: Oxford University Press, 2020).

12. 12 An exception is Martha Nussbaum, Not for Profit: Why Democracy Needs the Humanities (Princeton: Princeton University Press, 2010).

13. 13 For an external challenge not mentioned by Bush, see Tamson Pietsch, ‘Universities, Their Publics, and Climate Change, in Australian Universities: A Conversation about the Public Good, ed. J. Horne and M. A. M. Thomas (Sydney: University of Sydney Press 2022).