Hanne Weie Oddli, University of Oslo, Norway

To get a true picture of causal relationships in psychotherapy and counselling we need to approach the question from different angles by use of various research designs.

Why should we care about causality at all? Bertrand Russell, in his seminal essay ‘On the notion of cause’, declared that:

All philosophers, of every school, imagine that causation is one of the fundamental axioms or postulates of science, yet, oddly enough, in advanced sciences such as gravitational astronomy, the word ‘cause’ never occurs.… The law of causality, I believe, like much that passes muster among philosophers, is a relic of a bygone age, surviving, like the monarchy, only because it is erroneously supposed to do no harm.

(Bertrand Russell, according to von Wright, 1971, p. 35)

Much has been said about causality since (and before) then. Nevertheless, we frequently witness a general cautiousness about claims of causality in research. While acknowledging the extensive debates and wise arguments in favour of cautiousness, I think there are reasons to bother with causality. It has concerned philosophers at all times. We tend to organize events in causal relationships – they seem to be fundamental to us. This leads us to a first reason not to dismiss the question of causality: People, policy makers, and the various committees and councils who are given the responsibility for developing guidelines and criteria for sound practice will obviously ask for the effects of what we do. It goes with the ‘social contract’ that we are able to ensure people that therapy works, and to convey that in a way that is actually conceivable.

Second, journal editors and reviewers may call for it. If not necessarily intentionally, the call for causality often seems to be there as a fundamental premise for requests of methodological rigor. One of the reasons why the whole question of causality captured my attention at a certain point was that I began to reflect upon this seeming call – from journal editors, perhaps, or maybe just as well from colleagues who had their manuscripts rejected. The argument goes that the high-ranking journals favour analyses conducted by certain statistical models and preferably within an experimental design. The limitation sections of published articles frequently include cautions about claims of causality, highlighting that the studies are merely correlational and so on. Moreover, colleagues meet requests for quality assessments and requirements, such as control groups, even though it might not be in accordance with the particular design and aims of their study. So what is this all about? Does it reflect a mere confusion about proper quality criteria for research designs? Or is it perhaps a sound voice in there – a voice or a wish that I myself would agree with? Put very briefly, I think so. If this call for certain methods and designs first and foremost reflects a wish to identify complex processes and how incidents interact and work together, I am of course willing to embrace that, in itself. The question is, however, if these particular designs and methods are the only ones fit to develop knowledge about causal processes. If we accept that the call for those particular methods reflects a yearning for causal relationships, we will need to re-inspect the often non-articulated, taken-for-granted supposition that only certain experimental designs may provide this knowledge.

These are questions that my colleagues and I are dealing with in our recently published article titled ‘Causality in psychotherapy research: Towards evidential pluralism’ Here we describe three major philosophical perspectives of causality and outline some possible implications for research. These perspectives are: 1) The Humean conception of causality, which is a central philosophical underpinning for most experimental designs, such as randomized controlled trials (RCTs). Here, observed regularities of discrete events form the basis of causal claims. 2) ‘Dispositionalism’ and the power perspectives of various realist approaches. These emphasize intrinsic properties and contextual variations, highlighting causes as tendencies that depend on the interplay of a variety of factors to be actualized. 3) ‘Inferentialist’ and ‘epistemic’ approaches that advocate causal explanations in terms of inferences and linguistic interaction. Here, causality is placed at the level of human mind, implying that what counts as valid causal claims will depend on the discourses of the various communities of researchers.

The discussion is illustrated by the current situation within psychotherapy research, and the APA Policy Statement on Evidence-Based Practice. We conclude that clarifying causal relationships is an ongoing process that requires use of various designs and methods. We advocate a stance of ‘evidential pluralism’, acknowledging that each design has its potentials and limitations. For example, while some designs are fit for ‘difference-making’ others are fit for ‘causal production’. The first deals with whether causal relations exist or not, including the strength of the causal relationships. Such causal relationships may, for example, be identified by experimental and quasi-experimental designs. Causal production, on the other hand, deals with how causal relations work, that is, the causal mechanisms. These may be identified by, for example, single-case designs and qualitative methodology. It is important to note, however, that the distinctions between difference-making and causal production may be blurred, and some methods may fit in either one. The point is that the various difference-making and production oriented methods may be viewed as complementary sources of evidence to build robust, nuanced knowledge on causal relations of a phenomenon.

The main thrust of the paper is that no method is in a unique position to establish causality. We need empirical findings and rigorous theoretical arguments and discussions from various perspectives to complement the picture of causal relationships of a particular phenomenon. To ensure inclusion of the variety of these knowledge sources we will need to transcend the perspective of empiricism and the Humean conception of causality, not to exclude it, but to include approaches based in also other philosophical perspectives.

Oddli, H. W., Stänicke, E., Halvorsen, M. S., & Lindstad, T. G. (2022) Causality in psychotherapy research: Towards evidential pluralism, Psychotherapy Research, DOI: 10.1080/10503307.2022.2161433