they are being exercised or suffered; unemployed workers have the power to work even though they are not doing so now and iron is liable to rust even though some pieces never get the chance to. On this view then, a causal claim is not about a regularity between separate things or events but about what an object is like and what it can do and only derivatively what it will do in any particular situation. (Sayer, 2002, pp. 104–105)
An object cannot be reduced to being something in itself. It is only what it is by virtue of its causal powers and liabilities; these tell us something about what this complex object can do (Buch-Hansen & Nielsen, 2005, p. 25). In this under-standing of causality, the question of whether a causal power or liability of an object is activated or suffered is contingent – it is a possibility, but not a necessi-ty; we cannot predict whether it will happen. It is conditioned by other objects, their causal powers and liabilities, and whether or not these will meet, support, or block each other in potentially complex combinations (Sayer, 2002, p. 107).
With the transitive and intransitive dimensions and the Johari window, I focused on personal experiences, i.e. the access the persons studied and I as the research-er studying them have to the world, and the relationship between our (potentially different) understandings of the world and the world itself.
In contrast to this, the three domains of the world, and the complex objects in the real domain, explain more specifically how the ontologically existing world is, and how the sometimes unobservable and complex interactions between dif-ferent objects make it difficult to predict the future or to understand the world accurately based on observations of it.
Nielsen, 2012, pp. 284–285). In this thesis one of my aims is to move from the assumption that paid public-sector work in fact differs from volunteer third-sec-tor work to an understanding of some of the underlying mechanisms that cause this work to be seen and experienced as different on the surface. As we shall see later in the analysis, it seems that the activities conducted in volunteer scout work differ from the activities of the paid work of both Dorthe and Ditte in the sense that the scout activities in some ways do not seem to have an instrumental purpose outside themselves. This interpretation is not based on any single obser-vation, nor will I argue that this phenomenon actually can be directly observed as such. Ditte and Dorthe do not seem to deliberately frame their scouting activ-ities as having no instrumental purpose, but based on, among other things, their actions, their thoughts from interviews, and a structural understanding of the scout associations they take part in - this still seems to be the case - although this interpretation cannot be said to be the result of any (number of) observations.
Critical realist research looks for a special type of causation in terms of causal mechanisms:
[C]ausation is not understood on the model of regular successions of events, and hence explanation need not depend on finding them, or searching for putative social laws. The conventional impulse to prove causation by gathering data on regularities, repeated occurrences, is there-fore misguided; at best these might suggest where to look for candidates for causal mechanisms. What causes something to happen has nothing to do with the number of times we have observed it happening. Explanation depends instead on identifying causal mechanisms and how they work, and discovering if they have been activated and under what conditions.
(Sayer, 2000, p. 14)
Practical methodology: Retroduction
The question now becomes, how we can move from something observable down to the unobservable? A critical realist approach lies in the term retroduction.
Retroduction (or retroductive reasoning) is form a logical inference, that accord-ing to Buch-Hansen and Nielsen (2012, p. 304) is related to abduction3 and
3 Abduction and retroduction is seemingly by some researchers like Jespersen (2013) used as two terms describing the same concept, while others distinguish between the two as different but related terms. See Kringelum (2017, pp. 67–69) for a discussion of some of these differ-entiations.
which can perhaps best be explained by contrasting it with the more classic de-ductive reasoning. Dede-ductive reasoning is a logical movement, which involves reaching a conclusion through logical reasoning based on one or more state-ments or premises.
It can, as exemplified in Alrø, Dahl, & Schumann (2016, p. 180), consist of matching a specific case or entity to a general rule, thereby reaching a conclusion:
- Premise 1: All men are mortal (the general rule) - Premise 2: Enok was a man (single case) - Conclusion: Enok is mortal (result)
In retroductive reasoning, the process begins with the conclusion, which is al-ready given: there are observable phenomena in the domain of Empirical or Ac-tual (e.g. the notion that paid work differs from volunteer work), and the task of the researcher is now to infer which premise(s) in the domain of Real, could lead to this given conclusion. With this type of reasoning, the strict logical approach of deduction (as exemplified above) cannot be used and must be replaced with qualified guessing. Using retroductive reasoning, the task is to point out which necessary conditions (liabilities) and deep causal relations would most likely have to exist for this conclusion to be the case. This way of reasoning involves both creativity and using imagination as a researcher, and at times it includes using analogies or metaphors in explanations (Buch-Hansen & Nielsen, 2012, p. 304).
Sayer explains retroduction this way:
Wherever possible, we try to get beyond the recognition that something produces some change to an understanding of what it is about the ob-ject that enables it to do this. In some cases, such as that of gravity or the connection between a person’s intentions and actions, we know little about the mechanisms involved. What we would like in these latter cases, and what we already have in cases such as the conductivity of copper or the erosive power of a river, is a knowledge of how the process works.
Merely knowing that ‘C’ has generally been followed by ‘E’ is not enough:
we want to understand the continuous process by which ‘C’ produced ‘E’, if it did. This mode of inference in which events are explained by postu-lating (and identifying) mechanisms which are capable of producing them is called ‘retroduction’. In many cases the mechanism so retroduced will already be familiar from other situations, and some will actually be observ-able. In others, hitherto unidentified mechanisms may be hypothesised.
In the history of lay and scientific knowledge there are both cases where
such hypotheses have later been corroborated (e.g. viruses, capillaries) and where they have been rejected (witchcraft, heat as a substance). The philosophy of science cannot, of course, provide guarantees of success!
(2002, pp. 106–107)
Easton suggests using the question format: “What caused the events associated with the phenomenon to occur?” (2010, p. 123) when working with retroduction in analysis.
An example of retroduction and triangulation
When Dorthe instructs a group of scout children to make a sword out of duct tape, foam mats and conduit pipes, I can observe that she acts and communicates in certain ways: for example, she seems quite keen to get everyone involved; her actions seem to be mainly reactions to questions and initiatives from the children;
and there is nothing in the concrete interaction that points towards whether she is paid to be there or not. When interviewed, Dorthe tells me that she considers this volunteer work because it is unpaid associational work, but this will not lead me to argue that making duct tape swords with a group of children for no money is what volunteer work is, nor that volunteering can be reduced to the question of being paid or not. That would be reducing the idea of volunteering to being the same as these concrete observations, i.e. arguing that the transitive and in-transitive dimensions are the same. Instead, critical realism tells me that I must treat this as an example of what volunteer work can be (or the observable result of what volunteer work is in deep levels of the domain of Real), and try to infer what could make it be so. It tells me that what I am observing here is the (at least partially) unobservable interaction of a number of complex objects, where, for example, the intentions, understandings, abilities, and actions of the children in-teract with Dorthe’s intentions, understandings, abilities, and actions in complex ways. When Dorthe introduces the activity, she draws on her own structurally influenced background as a scout and her understanding of how a scout leader should act. This is not necessarily a conscious choice she makes, but as I shall argue in the following section on hermeneutics, she will inevitably draw on her own historical background when deciding what do to in the now. Thus, as one example of understanding volunteer associational work as a scout, I must both include aspects of not being paid as well as the implicit understanding of how scouts act, which will be partially observable in the way Dorthe acts and commu-nicates with the children. What causes this situation to happen the way it does is a lot more than the somewhat observable fact that Dorthe is not paid money to be there; this is also the case for what makes this volunteering, as I will argue
that any incidence of volunteering will be the result of a number of aspects com-bined, some of which are not accessible to the people doing the volunteer work.
After this introduction to critical realism as one of the two philosophies of sci-ence this thesis is based on, I will now move on to describing in more detail how the two types of work will be studied using a case-study approach based on the epistemology and ontology of critical realism. After this, we shall return to the philosophy of science of this thesis with Gadamer’s philosophical hermeneutics, which I use to describe the process of understanding or interpreting.
In the following section I will explain how case studies are used in this thesis. The approach is a combination of thoughts from Gary Thomas (2011), Erik Maaløe (1996), and Bent Flyvbjerg (2006). The section is structured around 10 character-istics of case studies that serve the purpose of describing both what a case study is and how this understanding of it applies to this thesis.