The object-side and the subject-side
Some statements by Thomas S. Kuhn have posed particular interpretative challenges for philosophers. One such is his view that 'though the world does not change with a change of paradigm, the scientist afterward works in a different world' (Kuhn 1970, 121). With regard to this, Paul Hoyningen-Huene reconstructs Kuhn's philosophy of science very interestingly in his Reconstructing Scientific Revolutions (1992). A particularly valuable distinction that Hoyningen-Huene makes is between two different kinds of worlds: the world-in-itself and the phenomenal world. The former is unknowable and purely 'object-sided', while the latter is constituted by the object-sided world-in-itself and by 'subject-sided' moments originating with an epistemic subject (Hoyningen-Huene 1993, 31-42; see also Devitt 1997, 72, 156-157). Epistemic subjects are thus co-constitutive of the 'phenomenal world', which is 'the scientist's world' and which changes in a revolutionary scientific change. The world-in-itself in turn contain 'no moments on the side of epistemic subject' (Hoyningen-Huene 1993, 33) and is thus independent of these subject-sided moments.
While I do not want to commit to the existence of the Kantian unknowable world an sich, I find the notions of 'object-sidedness' and 'subject-sidedness' extremely useful as parameters for indicating the origin of our views of the world. The distinction connects well to the discussion concerning the notions of objectivity above. If one managed to construct a purely object-sided view it would be independent of any subjectivity of the constructor; it would be entirely of the object. This would correspond to the ideal of ontological objectivity. It is worth citing Kuhn here on the Cartesian epistemological tradition: it assumed that 'observations ... themselves are fixed by the nature of the environment and of the perceptual apparatus', which is seen as universally shared. No cultural, psychological or other subjective factors affect the process (Hoyningen-Huene 1992, 37; Kuhn 1970, 120). On the other hand, a view that is a total fantasy has nothing to do with the subject- independent object world. In this case it would be purely subject-sided or, in brief, a subjective view.
In the early stages of modern historiography toward the end of nineteenth-century and the beginning of the twentieth-century, attaining objectivity was seen as an achievable prospect. Ranke's influential request to write history wie es eigentlich gewesen may be taken as exemplifying the desire for ontological objectivity (cf. Iggers 1962, 1973). In recent years, objectivity as an aim in historiography has been a subject for criticism and has even been seen as harmful by some accounts. Often the explicit target of criticism has been objectivity as neutrality (e.g. Novick 1988; Newall 2011). Typically the question of whether an objective account is possible has been put very categorically. One tries either to prove that historiography is/can be an objective science or otherwise one succumbs to a position that historiography is inexpugnably subjective for some reason. This kind of dichotomical thinking can, perhaps surprisingly, be found even among postmodernists. For example, for Keith Jenkins and Elizabeth Ermarth there is only subjectivity, that is, the validity of any view is only the validity of its creator for its creator. Pihlainen argues that 'all stories and ways of structuring material' are 'equally imaginary', which necessitates abandoning 'the last illusions of history as somehow "objective"': 'Objectivity as it has traditionally been understood within history is simply not an option' (Pihlainen 2013b, 516). While the critique of the traditional take on objectivity in historiography may well be justified, it is nevertheless ironical that although postmodernists are the most vocal in talk against dichotomies and dichotomical thinking, they are the ones who seem to commit to these most strongly (that is, postmodernist-modernist historians, subjective- objective, etc.). In the case of objectivity, they seem to understand it to mean some very strong notion of ontological objectivity, and when that is judged not to be universally applicable, the argument ends with the conclusion that we are confined within our subjectivity.4
More specifically, the narrativist and postmodernist argumentation takes the following form. 'Narrative' is recognized as the main knowledge contribution of historiography and is identified as a holistic entity that possesses qualities not found in the research object, that is, in the past. It follows that ontological objectivity is impossible and that narrative contains irreducibly subject-sided properties. And because narra- tivity defines historiography as a scholarly practice, historiography is indispensably subjective. Reference to facts to provide the bedrock for narratives does not help. First, the talk of 'facts', including the border between 'fact' and 'non-facts' is problematic. More importantly, the relation of 'facts' to the higher-order form 'narratives' is unclear, to say the least. As we have seen in Chapter 3, as a consequence of being identified as literary entities, narratives are analytical, unfalsifiable and immune to other narrative challenges.
An alternative view of the relationship between objectivity and subjectivity is based on the rejection of 'narrative essentialism', as it was called in Chapter 5: narrativity does not define historiography, that is, it is not an essential feature of a historical presentation; instead, reasoning and argumentation characterize historiography. As a result of this change of perspective, the form of presentation does not deem historiography subjective (even when we focus on texts and complete works of historiography), holistic and automatically immune to empirical challenges. Further, the form does not make it impossible that some features are grounded in the objective-sided world. With regard to the presentations of historiography and their components, one needs to be more circumspect than to assume that they are all of one and the same kind.
I would like now to suggest that one learn to live with the 'sliding scale of objectivity and subjectivity'. In others words, one would talk about the degrees of objectivity and subjectivity that a particular historical presentation possesses. And the degree of objectivity and subjectivity must be investigated case by case. Moreover, my view is that good historiography may possess objectifying features in both ontological and justificatory senses of 'objectivity.' First I show how the notions of 'objectivity' and 'subjectivity' apply to historiography in the ontological sense. The application of 'objectivity' in its justificatory meaning will be explicated below, when the topic of rationality is discussed.
I go back to the distinction between subject-sidedness and object- sidedness. The distinction differentiates between those entities that derive from an epistemic subject and those whose ontological standing is independent of human cognition. If all knowledge were derived from the object-side, one would have a purely objective account. Note however that this may not necessarily be the most desirable outcome, even if it is possible, as the result might not be very interesting historiographically. I will elaborate on this below. Note also that the distinction between object-sidedness and subject-sidedness has the advantage that one does not need to commit to the metaphysically problematic notion of 'facts' and their separation from 'non-facts', as the narrativists, postmodernist and also historical realists typically seek to do.
It is best to first mention an example of the object-sidedness of knowledge that one can find in books of history. The simplest case would be a claim that refers to a person and a state of affairs that hold with regard to that person. For example, 'Stalin owned a gun' or 'President Urho Kekkonen was born in 1900'. I do not know whether Stalin owned a gun, but if he did then this sentence is true due to object-sided factors, and if he did not then the sentence is false. We may of course never ascertain this, but that is a different matter.
The reader may already be thinking that books of history only very rarely make cases for these kinds of simplistic objectifying statements, although such statements are, of course, part of a historiographical text. The thought that historians form synthesizing interpretations or other integrative views and concepts immediately implies that historical knowledge contains subject-sided elements. Indeed, insofar as the central historiographical craft is the production of texts, as the narrativist philosophy of historiography has emphasized, I take it as a (new) default position that historical knowledge is at least in part subject-sided.
What kinds of elements may then be said to contribute to the subjectsidedness of a historical presentation? There are four main categories that I can think of: (1) the absence of reference; (2) the postulation of a nominal categorizing principle; (3) the postulation of narratively construed causal relations; and (4) the postulation of meaning or sense to the past. In what follows, I will introduce an example of each.
- 1. The absence of reference is common with colligatory and metaphorical concepts. Any entity to which a linguistic term can refer is 'reference'. Most typically it is an individual, a person, to whom a name refers, but it can also be any other object insofar as that object exists independently of linguistic postulation, as some specific gun of Stalin's. Normally, reference is to something materially existing but some philosophers also accept abstract entities as objects that can be referred to. For example, a colligatory expression is an arrangement of historical data into larger wholes and does not itself refer. Arguably there is no such entity or even process as the 'Renaissance' that is given and exists prior to historians' practice. This was discussed extensively in Chapter 6. And if there is no reference, then the 'Renaissance' is a constructed and subject-sided historiographical entity.
- 2. The postulation of a nominal categorizing principle means a situation in which a historical category has been formed and a number of lower-order entities are subsumed under the category concept but the organizing principle is not given in the object-sided world. It is not 'natural' in any sense. When would such a principle be 'given'? The best example would be the case of natural kind categories. Elements are typical examples of such. Why are all samples of gold of the same kind, that is, why do they belong to the same natural category? This is because there is an essential property, perhaps definable by the possession of the atomic number 79, that they all share and that is allegedly objectively given in the sense that it would have been there even if no human had ever existed. Now let us compare this case once again to an example in historiography. The 'Renaissance' can be seen to organize a very diverse set of objects, paintings, furniture, dressing styles, literature, thinkers, etc. with the consequence that there is no shared property across all the objects that are organized under the concept. This would be a straightforward case of subject-sided organization. But even if one postulated a narrower definition of the 'Renaissance' so that all instances of the 'Renaissance' need to share some kind of artistic style, for example, it is still the case that this feature is not natural in the sense that it would give us a self-evident category devoid of the subjective choices of the historian.
- 3. The postulation of narratively construed causal relations means that a historian postulates a complex narrative of historical events that together form a historiographical object, such as a process. E. P. Thompson's (1991) thesis of the emergence of the English working class could be seen as an example of this. It unites a very large set of historical events, including methodical religious ceremonies at the end of the eighteenth-century, the meetings of London-based learned societies at the beginning of the nineteenth-century, illegal activities after 1815 in Britain, changes in the material and economic situation of skilled workers, etc. One can hardly argue for the existence of explicit object-sided causal connections between all these events, even if these events are central to Thompson's thesis on the birth of the English working class. It appears that causal relations are absent in the object world and the links, causal and other, between the events are postulated by the epistemic subject, that is, by the historian.
- 4. The final case deals with something that especially postmodernists are keen on emphasizing. That is, that historians add some kind of meaning or sense to the past, which cannot be said to be a property of the past itself, but derives, for example, from literary and cultural tropes and emplotments. A good example is Clark's (2012) portrayal of the main players in the development towards the Great War as 'sleepwalkers'. 'Sleepwalking' does not seem to be ingrained in the historical world itself, but is indeed a particular interpretation by a particular historian. In other words, the sense of sleepwalking is a subject-sided addition and a historiographical construction.
My suggestion is that practically all historiography needs to be considered as both subjective and objective at the same time; we should thus talk about the degrees of objectivity and subjectivity of a particular work depending on the specific combination of objectifying and subjectifying elements. It is clear that all books impose meanings on the past and contain straightforward referential statements, metaphors, colligatory notions, narratively construed causal claims, etc., the elements of which all relate to the research object in different ways.
In order to illustrate how subjectifying and objectifying elements can combine in a work of history, let me return to the idea that a rationally warranted colligatory notion must exemplify the historical data it applies to (as discussed in Chapter 6). The implication is that the descriptive content of a colligatory expression must appear to be appropriate with regard to historical evidence, which in turn entails that what is colligated, lower-order statements, must refer to the historical reality. If the release of the prisoners from the Gulag is a colligated feature of Khrushchev's 'Thaw' then, on some fundamental level, the expression must refer to actual 'given' prisoners and the locations at which they were held. They constitute in this sense the objective grounding of the notion of 'the Thaw', which is nevertheless a colligatory and thus subjective construction to a high degree. And it is worth emphasizing this point once more. One should not see the notions of 'subjective' and 'objective' as exclusive; not even in the case of ontological objectivity. The subjectivity that the historian imposes often means the integration of several events, which gives them some specific significance. If they were not object-sidedly grounded at all, they would be entirely fictional entities and could not possibly be informative concerning the research object.
The scale between subjectivity and objectivity is sliding. At one end, we might perhaps imagine a book whose only point is to prove, say, that President Urho Kekkonen was born in 1900. Perhaps the year of his birth had been a mystery for a long time, or there had been lots of conflicting sources and other information on it to provide the incentive for a historian to straighten it out (in actuality, there is not). Given that there really was an Urho Kekkonen and his birth occurred in that year, the truth-functional standing depends on the object. (Now, of course the Julian calendar imposes a culturally loaded standard, which means that not even this statement is 'purely' object-sided). The problem, if one wishes to call it such, and the narrativist insight, as it was called in the first chapter of the book, is that historiography typically goes beyond and above these kinds of simple referential statements and puts forward synthesizing theses concerning the past, the consequence of which is that one loses the possibility of attaining 'pure objectivity' from the very start. The one who sticks to grassroots-level expressions will end up impoverishing historiography and removing its most interesting expressions. Logical positivists tried to purify scientific language in the 1920s, but that attempt ended in failure. I would like to suggest that we would need very strong reasons for trying something like this again because the rationale would be external to historiography. That is, the motif for the simplification of well-functioning historiographical language would stem from the concerns of philosophy and philosophy of science.
It is time to end this section on ontological objectivity with an illustration of the sliding scale of objectivity. This is meant as a hermeneutical tool to illustrate the main idea. All the examples located on the axis are texts. Alice's Adventures in Wonderland by Lewis Carroll has hardly any referring statements to actual persons, their properties and the states of affairs. It is a work emanating mostly from the subject-side, although not necessarily entirely.5 Stephen Clark's The Sleepwalkers contains many referential statements, but ends with a synthesizing thesis on the process towards the Great War as 'sleepwalking'. It has other narrative and metaphorical parts that could possibly not literally 're-present' the past. The third example is an imaginary study, whose central thesis is that 'Urho Kekkonen was born in 1900'. Whether or not he was born in this year
Figure 10.1 The axis of subject-sidedness and object-sidedness
depends on the external states of affairs in the past. That is the closest it is possible to get to an objective work of historiography.
The reader has probably noticed the inverse relationship between object-sidedness and the originality of these works. Any book of historiography that intends to prove that someone owned some object or was born in a specific year is not very exciting; nor is it very interesting in normal circumstances. It can be very secure knowledge of the past, but very trivial and uninteresting at the same time. But if one argues that the main players in the process towards the Great War were 'sleepwalkers', then the claim is much more interesting. Or if a historian argues that the Finns - or at least a significant part of Finland's elite - self-censored themselves and profoundly changed the Finnish political culture of the 1970s under pressure from the Soviet Union (the thesis of 'Finlandization'), the historian makes an interesting statement. Historians are thus faced with the following dilemma with regard to objectivity and subjectivity: The more a historian is willing to state about the past, the less objective but
Figure 10.2 The inverse relationship between originality and objectivity in historiography more original the historian's account will be; and the more the historian desires objectivity, the less he or she is able to state and the less original the account will be.
It is necessary to emphasize that I do not see any problem with the inverse relationship between originality and objectivity. It just reflects the nature of historiography. Once again, one would not wish to see firmly objective but minimally interesting and insignificant theses, but neither would one want totally outlandish claims with near-absent objective-side grounding. Historiography, denoted by the black arrow, is located somewhere between these two extremes, as the figure shows. In different ways this view connects to those of some predecessors. It is well known that Popper argued, in his demolition of inductivism in the history and philosophy of science, that scientists should attempt to falsify rather than to prove scientific hypotheses as correct. The bolder a scientific hypothesis is, the higher degree of falsifiability it has. It is thus easy to falsify, but if it is stands the tests, it amounts to an extremely interesting and explanatory hypothesis: 'every interesting and powerful statement must have a low probability; and vice versa: a statement with a high probability will be scientifically uninteresting, because it says little and has no explanatory power ... as scientists we... seek explanations; that is, powerful... theories' (Popper 1989, 58). The issue of falsification is not as simple as Popper thought but his view nevertheless expresses something essential about scientific practice. What Einstein did was not to put forward secure observational statements on the environment in his near proximity but risky conjectures about the universe. Obviously, this analogy should not be overstretched in the context of this book since I do not think that there is a role for clear-cut falsifiability and certainly not for verisimilitude in historiography, as Popper assumed there is in science. Other predecessors can be found from nearer the field of this thesis. Dray pointed out that the idea of originality in historiography leads directly back to the discussions on the possibility of objectivity in historiography (Dray 1991, 173). Martin argues that our best and most fully developed interpretation will always occupy a halfway house between literature and science. That is not necessary, writes Martin, but 'that is the way we like our historical interpretations' (Martin 1993, 49). Martin concludes that 'perfect objectivity' may not be desirable, because we simply want something more in historiography. Finally, Ankersmit suggested at the end of his Narrative Logic that the 'the essential duty of the historian is to be original and to refrain as much as possible from repeating what his predecessors in the investigation of a particular topic have said' (Ankersmit 1983, 220). But again there are differences too. Ankersmit's comparison of 'narratios' to propaganda fails because historiographical theses are answerable to rational evaluation. More importantly, Ankersmit links objectivity in historiography to scope so that 'The most objective narratio, the narratio having the widest scope, is the least conventionalist, the most original narratio' (218).6 I think it is exactly the other way round, terminologically at least: an account that owes most to the subject-side is bound to be the boldest and most original.