Critical Realist Ontology & Epistemology

Critical realism is one extension to a dominant paradigm in social sciences research 0D[ZHOOS, scientific realism, which positions “the view that entities exist independently of our theories about them” 3KLOOLSV. A critical realist researcher believes in the existence of a real-world independently from our knowledge about it, and posits that one can apprehend only a part of real-world based on the limitation of human knowledge. Further, the mental and physical SKHQRPHQDDUHSDUWRIWKLVZRUOGLWDOVRLQYROYHVWKHDFWRUV¶LQWHUYLHZHG experts)

perceptions about a given situation and the researcher’s mental models (Maxwell, S . Consequently, critical realists share the positivist ontology and constructivist epistemology 0RUDLV. Ontologically, a critical realist believes in the existence of world independently from our knowledge about it. While epistemologically, critical realists believe that our understanding of the world has been affected by the already available descriptions and discourses. Thus, the creation of our new knowledge about the world is always theory laden. The critical realist ontological and epistemological perspective “instead of revealing reality through subjective constructivism,” lets the reality itself guide the research process $QGHUVRQ .UDJKS-.

This research understands that the world is characterized by emergence, in which physical and behavioral (social) objects are participatory through their structures DQG FDXVDO SRZHUV FRPSDUH 6D\HU 2ntologically, a phenomenon is constituted from real, actual, and empirical domains of stratified reality (Figure , PRGLILHGIURP6D\HUDQG0RUDLV+HUHLQ³WKHUHDO´GRPDLQRIWKLV stratified reality consists of structures and causal power of objects and entities.

This real domain of reality is directly unREVHUYDEOH +RZHYHU LWV existence and actuation are irrefutable. The structures from the real domain actuate causal mechanism under given contextual settings. These mechanisms are how certain entities exercise their causal power and create observable events in the actual domain of reality. The relationship between causal powers (or mechanisms) and their effects is QRWIL[HGKRZHYHURQO\FRQWLQJHQW6D\HUS,QDJLYHQ context, these causal mechanisms trigger in the real domain and engender observable events leading to the experience of a particular phenomenon (Sayer, . In another context, a mechanism acts differently to produce different outcomes (i.e. observable events). Inversely, a single event can be the outcome of FRPSOHWHO\ GLIIHUHQW FDXVDO PHFKDQLVPV %OXQGHO S (DVWRQ 6D\HU7KHVHFDXVDOPHFKDQLVPVDUHQRWGLUHFWO\REVHUYDEOHZKHUHDVWKHLU generated events are observed as the subjective experiences of humans, and constitute the empirical domain of reality (Figure ).

Accordingly, this research posits that an organizational entity actuates the causal mechanisms of capability development in the real domain of reality. The actuation of these mechanisms is organizational business context dependent, as are the outcomes (organizational capabilities) generated by the causal mechanisms.

Therefore, an organization’s capabilities always remain idiosyncratic in their details (its dimensions, routines, and performance outcomes). This research, therefore, understands that neither the capabilities development process nor do the developed capability in one organizational context can be perfectly imitable as

“best practices” to another context (OrlikowskL)XUWKHUPRUHLQDQRWKHU

organization with a similar business context, dissimilar causal mechanisms can actuate to develop the same capabilities through a different development path.

+HQFH WKH FDSDELOLWLHV ZLWK VLPLODUPDWXULW\ OHYHOV XVXDOO\ PHasured based on capability best practice models, shall correspondingly result in different performance outcomes. Consequently, a capability, developed through different paths due to actuated causal mechanisms and contexts, can take alternative roles in the organizational development (either as ordinary capabilities or dynamic capabilities) and contributes to its business performance accordingly.

Figure 4. Critical realist’s world view of stratified reality

The actuated causal mechanisms do not become obvious to the participants contributing to the development of organizational capabilities. Because these participants can only observe the temporally sequenced events as the generated outcomes of causal mechanisms. Their observation of these events depends on (their domain specific) subjective knowledge about the capability dimensions and routines (discussed in section ). Accordingly, organizations’ (expert) resources can imperfectly interpret the causes of these events. Meanwhile, their interpretations are based on the limited exposure to certain capability routines and performance outcomes. The critical realists aim to further clarify the reality by explaining the capability development causal mechanisms through observed events and the experience of expert resources in the case organization (Table ).

Accordingly, a retroduction-based process theory approach was adopted to explain this capability development process in the case company (The Company Alpha) and its two business divisions.

Table 1. Organizational capability development ontology [modified from Causal Mechanisms Yes

Events Yes Yes

Experts’ Experiences Yes Yes Yes

Critical realism has the capacity to explain empirical events through their causal processes 20DKRQH\ 9LQFHQW . +HUHLQ WKH FDXVDO PHFKDQLVPV DQG actuating contexts are LQH[WULFDEOH SDUWV RI WKHVH FDXVDO SURFHVVHV $FNUR\G .DUOVVRQ 0D[ZHOO S ZKLFK WHPSRUDOO\ DQG VSDWLDOO\ WLH-up the H[SHULHQFHGHYHQWVLQWRLQYHVWLJDWHGSKHQRPHQRQ0D[ZHOOSUHIHUUHG it as a process-theory approach toward building an explanation based on

“theoretical propositions” 20DKRQH\ 9LQFHQWS According to Miles DQG+XEHUPDQS-), such an explanation brings “local causality” by linking the events with the processes leading to a specific outcome, and it offers superior value to scientific research. This can be distinguished from positivism and empiricism, which remain limited to only outlining a phenomenon based on regularities and variance in empirical experiences 0D[ZHOO20DKRQH\

9LQFHQW . In practice, explaining the above-described local causality (involving context and mechanism) requires close association with the phenomenon 6D\HU . Accordingly, the later described methodological selections in this PhD research reasonably convene the ontological and epistemological choices discussed in this section (detailed below).

2.1.1 Retroduction

Determining the mechanisms responsible for organizational capability development process proved a demanding task [as it was cautioned by Pettigrew,

@%HFDXVHLWUHTXLUHGDQXQLQWHUUXSWHGLQWHUSOD\EHWZHHQWKHFDVHGDWDDQG the extant literature. Essentially, such an approach is referred as retroduction which ensures the refurbishment of conditions (the context) by which the temporally sequenced events (in time and space) led to the investigated phenomenon %OXQGHO S : the development of project portfolio management capability (PPMC). In addition, retroduction enable a plauisble explication of causal mechanisms through transfactual generalization (Morais, S Therefore, contrary to the induction or deduction based logic, retroduction is a “mode of inference in which events are explained by postulating

(and identifying) mechanisms which are capable of producing them” (Sayer, S.

Retroduction entailed imagining an ‘all-inclusive’ model of mechanisms, “which if it were real, would account for the phenomenon” %KDVNDU S YLL of capability (PPMC) development in the research case context. Accordingly, this PhD research adopted retroduction as DREIC model of theoretical inquiry to guide the research process (compare%KDVNDUSYLLDVIROORZV

[D] Describing (theoretically) the pattern of identified events related to the capability development process.

[R] Reproduction of logically disjunctive explanations about the alternative capability development paths.

[E] Eliminating the less compelling explanation through interpolation of applicable theories and case data.

[I] Identification of causal mechanisms that actuated during the data collection period in the research case.

[C] Continuously correcting the earlier findings along the data collection and analysis process.

Simply stating, this PhD research has operationalized retroduction as a meta-process to develop knowledge about capability development (compare: Easton, ,QSUDFWLFHLWUHTXLUHG adding appropriate theoretical insights to the case data +RGJNLQVRQ 6WDUNH\20DKRQH\ 9LQFHQWS. It resulted in gradual enhancements to the researcher’s knowledge and required continuously revisiting the theory and the recorded events through interpolation $QGHUVRQ .UDJK. Earlier in literature, 7HHFHDSKDVDGYLVHGH[DPLQLQJ organizational capability development through “introspection and observation”.

Such an examination obliges combining the deductive and inductive logics of qualitative inquiry: a critical realist retroduction-based approach (compare:

(DVWRQSIt entailed an extended engagement of the researcher with project portfolio management capability (PPMC) development process in the UHVHDUFKFDVHFRQWH[WFRPSDUH(ULNVVRQ)RVVHW DO*UDQW 9HURQD +HOIDWHWDOS-7HHFH:LQWHU

Figure 5. ‘All-inclusive’ model of mechanisms for PPMC development in the case company

For readers’ understanding, it is important to specify that this ‘all-inclusive’ model (Figure 5) of capability development reflects the researcher’s approach towards data collection. Because retroduction required imagining all those mechanisms (compare %KDVNDU S YLL which could have been responsible for the heterogeneously developing PPMC at TCA. Through introspection and case observations this model kept refining along with the researcher’s knowledge about the phenomenon. Such a retroduction-based approach enabled the researcher to reach to a generalized framework for investigating organizational capabilities (Figure /DWHU WKDW JHQHUDOL]HG IUDPHZRUN ZDV SRSXODWHG WR UHQGHU 330&

through literature synthesis (Segment II) and presentation of the case results (Segment III).