Introduction

Research disciplines have many intellectual and administrative benefits, and will likely continue to be central to knowledge creation and innovation into the future as they have been throughout the past century at least (Jacobs 2014). Nevertheless, collaboration among experts with different backgrounds and training can also be beneficial, as it has been linked to improvements in innovation (Edmondson and Harvey 2018), creativity (Guyotte et al. 2015), performance (Sun et al. 2021), and impact (D’Este and Robinson-García 2023). Furthermore, there is an increasingly popular view that real-world policy problems cannot be addressed with knowledge from a single research discipline alone (Andersen et al. 2023; Bammer 2013). Some of the most pressing problems relevant to public policy in jurisdictions around the world, in areas such as poverty, homelessness, environmental protection, and public health, are inherently interdisciplinary. While a system built on traditional research disciplines might have its merits, there is reason to believe that interdisciplinary collaboration can contribute additional gains in knowledge and innovation.

However, there is still much that is not known about how to enable and support interdisciplinary research collaborations, especially for the formation of new collaborative teams. One particularly popular hypothesis about the formation of interdisciplinary teams involves potential cultural differences between disciplines, which many observers see as a significant barrier to collaboration (as opposed to material factors like funding opportunities—see Newman 2024a for example). Studies have pointed to differences in language (Austin et al. 2008: 562), methodologies (Barry et al. 2008), and values (Brister 2016: 88), among other things, all of which increase the “transaction cost” of working with researchers from other disciplinary backgrounds (Beers and Bots 2009). Many authors argue that research disciplines can become “tribal” (Siedlok and Hibbert 2014: 204) or practice “protectionism” (Lattuca 2001: 26), staunchly defending their research territory in a way that is antagonistic to interdisciplinary collaboration. The most basic, and most widely discussed, of these tribal divisions is between science and arts (Brown et al. 2015; Gardner 2013; Lach 2014). However, while this might sound intuitive, and despite much discussion and debate on this subject over many decades, there is really very little empirical evidence that can confirm many of these claims (Newman 2024b).

In light of the above, this study asks three main questions: first, is there a demonstrable cultural divide between science-based and arts-based academic research disciplines? Second, if such cultural divisions exist, do they present barriers to interdisciplinary research collaboration? And third, if cultural divisions between science and arts can be shown to hinder interdisciplinary collaboration, what are institutional leaders doing or planning to do to mitigate the effects? The assumption throughout is that interdisciplinary collaboration is desirable—at least in the eyes of leadership at universities and research institutions, if not more widely (Brown 2018; Leahey and Barringer 2020).

This article reports on a case study conducted at a large research-intensive university in Australia, in which cultural barriers to interdisciplinary research collaboration were investigated. Monash University is one of Australia’s top-tier “Group of Eight” research-focused universities. It is home to researchers spanning a wide variety of arts, science, engineering, medicine, and humanities disciplines, and is well known nationally and internationally for world-class research. At the time this study was conducted, Monash was ranked within the top 100 universities in the world on a number of independent higher education ranking schemes. It is among the biggest universities in Australia in terms of number of students and courses taught, number of researchers employed, and dollar amount of external research grant funding obtained per year.

Monash University is a good candidate for a representative case study on interdisciplinary research collaboration. There is nothing particularly unique about Monash’s research or teaching output, either in the Australian or the international context. Monash does not specialize in any particular area of research, and has good infrastructure and intellectual capital in a wide variety of research areas spanning both the arts and sciences. It has an excellent national and international reputation and attracts large numbers of international students each year—but not in significantly greater proportions than equivalent universities in other jurisdictions, either in Australia or elsewhere. There is also ample opportunity for researchers at Monash to collaborate across disciplines. Meanwhile, Australia has a long history of mythologizing the intellectual divide between arts and sciences (Barnes 2023).

This study is based on a survey of 430 research academics, and interviews with 14 senior leaders, at Monash University. Survey responses, representing a broad range of ages, career stages, and research disciplines, revealed significant differences between science-based researchers and arts-based researchers, which suggests that there might be a genuine arts-science cultural barrier to interdisciplinary collaboration at the university. In particular, science-based researchers reported reluctance to engage with arts-based researchers and their research, while the reverse was not apparent. In-depth interviews indicate that the issue, while pervasive, has not yet captured the attention of university leadership.

Taken together, this study presents a case in which cultural divisions between researchers could pose a barrier to interdisciplinary research collaboration. The nature of the views held by survey respondents, and the invisibility of the issue to the university’s leadership and general community, suggest that a coordinated strategy would be required if the problem were to be confronted. Furthermore, if this problem is not unique to Monash University, then there are broader implications for public policy in general, and more attention would need to be paid to general efforts to improve understanding and cooperation between the arts and sciences.

Interdisciplinary research and challenges to collaboration

Operationalizing the concept of interdisciplinarity for empirical research is a challenging task (Newman 2024b). A widely-cited report by the US National Academies defines interdisciplinarity as “research by teams or individuals that integrates information, data, techniques, tools, perspectives, concepts, and/or theories from two or more disciplines or bodies of specialized knowledge” (Institute of Medicine 2005: 26), but many authors find this definition to be incomplete, and further attempts at pinning the concept down continue to be undertaken (e.g. Aboelela et al. 2007; Klein 2017). Many studies describe three different levels of interdisciplinary activity, ranging from “multidisciplinarity”, in which knowledge from different research disciplines is used in parallel, to “interdisciplinarity”, in which some boundaries are crossed but where the original disciplines are still apparent, to “transdisciplinarity”, where disciplines are blended to create new fields of knowledge, such as astrobiology or sustainability science (Austin et al. 2008; Gibson et al. 2019). However, these terms are highly subjective and sometimes confusing: for example “transdisciplinary” is often used in a completely different sense, to mean research that is co-produced with end users (e.g. Frodeman 2011). Furthermore, the important distinction between interdisciplinary research, which can be produced by single researchers working across disciplines, and interdisciplinary collaboration, where researchers from multiple disciplines collaborate (but do not necessarily produce interdisciplinary research), is often ignored (Newman 2024b).

In light of these challenges, this study took a pragmatic approach to operationalization. The purpose here is to investigate cultural barriers to interdisciplinary collaboration, that is, collaboration between researchers with backgrounds in different research disciplines. In order to minimize the conceptual work needed for respondents to complete a survey on this topic, a very basic definition was offered: interdisciplinary collaboration was defined, in the survey and in interviews, as working on a research project with partners who are affiliated with a different research discipline than one’s own primary affiliation. All references here to “interdisciplinary research” and “interdisciplinary collaboration” should be understood in this context.

There is a widespread set of beliefs that cultural differences between research disciplines create obstacles to interdisciplinary research collaboration (Brister 2016). The most prominent of these beliefs is that researchers from different disciplines speak in unique “languages”, using terminology or jargon that is unintelligible to outsiders (McCoy and Gardner 2012: 46), and that researchers must learn to speak one another’s language if they want to collaborate (Klein 2014). Accordingly, a great many authors argue that communication between researchers with different research backgrounds is inherently difficult, and that this communication barrier hinders interdisciplinary collaboration (Austin et al. 2008: 562–563; Fraser and Schalley 2009; Szostak 2014; Winowiecki et al. 2011: 75).

Other cultural barriers to interdisciplinary collaboration have also been noted. Numerous authors, for instance, claim that disciplines use particular methodologies or epistemologies, and therefore interdisciplinary collaborators may not be familiar with the tools of the trade used by their counterparts (Gardner 2013; Lach 2014; Pedersen 2016: 4). Similarly, disciplines have been ascribed their own sets of values, which can produce diverging research objectives and analyses (Robinson et al. 2016), such as when the values of conservation biologists, who want to protect biodiversity, might conflict with those of social scientists, who might instead prioritize the needs of local human communities (Brister 2016). Values divergence at this level would make interdisciplinary collaboration extremely challenging.

If research disciplines have their own unique languages, tools, and values, it is reasonable to expect that members of particular disciplines might engage in some form of tribalism. At the more benign end of the spectrum, disciplinary tribalism might simply be a sense of identity or belonging in a community of like-minded colleagues with common research interests (Lattuca 2001: 36). However, disciplinary tribalism might also produce negative effects, especially if it results in a bias toward a particular way of thinking or an unwillingness to explore research questions, methodologies, or results that contradict an established canon (Robinson et al. 2016). At the far end of the spectrum, tribalism could become a form of protectionism; discipline members might feel threatened by other disciplines, and they might engage in protectionist activities, guarding job markets, research funding, and PhD student placements against encroachment from perceived outsiders (Jacobs and Frickel 2009: 55).

The most widely reported form of disciplinary tribalism is the alleged cleavage between science and the arts. At least since C. P. Snow (1959) famously wrote about the “two cultures”, scholars have debated the existence and significance of this divide (e.g. Andreasen and Ramchandran 2012; Darbyshire 1999; Haste 2016; Leavis 1962; Massey 2019). One major source of contention in this debate is the vague but historically pervasive suggestion that the “hard” sciences (Gardner 2013)—physics, chemistry, biology, mathematics, medicine, and engineering—are more rigorous, and perhaps therefore more legitimate, than other disciplines (Brown et al. 2015; Goldbard 2015; Sorell 1991). If the perception that some classes of research are more rigorous than others currently exists among researchers, this could also present a significant barrier to interdisciplinary collaboration.

Philosophers have tried to define “science” for a very long time, at least since the time of Aristotle (Cohoe 2022). And yet, separating science from art in any definitive way has proven quite difficult. Weber described science as a search for facts that progresses over time (in that new facts emerge that disprove earlier facts), as opposed to arts, in which the product of one era cannot be said to be superior to the product of another (Weber 2013 [1917]: 137–140). However, Weber, a sociologist, made no distinction in his formulations of science between science that relates to the natural world (e.g. physics) and the sciences of human society (e.g. economics), the latter of which we might now assign to the arts (Weber 2013 [1917]; OECD 2015: 56).

In his essay on the two cultures, Snow referred to “humanists” and “scientists” but did not offer a precise definition of either (Snow 1959). More recent writers distinguish between “science and technology” on the one hand, and “humanities, arts and social sciences” on the other (e.g. Massey 2019). The Frascati Manual, a gold standard guide for classifying research and development, separates “the natural sciences and engineering” from “the social sciences, humanities and the arts” (OECD 2015: 54–55). However, again, precision is often lacking, and it is not always clear, for instance, why computer science should be joined with the “natural” sciences when it is about as far from natural as is possible, while sociology should be considered an art when it has a long tradition of the use of scientific method (Giddens and Griffiths 2006). The Australian and New Zealand Standard Research Classification, which divides research disciplines into specific categories for the purposes of funding allocation (e.g. 37 Earth Sciences, 45 Indigenous Studies), has been since 2020 organized in alphabetical order and makes no distinction between science and arts (ABS 2020).

And yet, divisions between research areas that are perceived to be “science” and those that are deemed to be “arts” are manifest. Monash University has a Faculty of Science, which includes biology, chemistry, and physics, and a separate Faculty of Arts that houses history, languages, and social sciences like criminology. Many Australian and international universities are organized similarly: Oxford, for instance, is divided into Medical Sciences, Mathematical, Physical and Life Sciences, Humanities, and Social Sciences; Princeton University is divided into Engineering, Natural Sciences, Humanities, and Social Sciences; McGill University has a Faculty of Arts that includes Art History, Economics, Linguistics, and Political Science, and a separate Faculty of Science that includes Biology, Chemistry, and Physics; the National University of Singapore has a Faculty of Arts and Social Sciences and a separate Faculty of Science. Even those universities where disciplines such as chemistry and political science are grouped in a single faculty (e.g. Yale University) use titles such as the “Faculty of Arts and Sciences”, which further reinforce the distinction. In other words, even without precise definitions, it may be that divisions between arts and sciences are somewhat intuitive.

This lack of definitional precision presents a dilemma for empirical research. Numerous authors (for example, those cited above) describe tension between arts and science disciplines, but no widely accepted definition of either side of this rift is on offer. Rather than attempt to settle a longstanding debate, here I will follow convention in delineating between research disciplines that are widely perceived as belonging to “science” (medicine/nursing/health sciences, natural science/maths/information technology, and engineering) and those that are often referred to as the “arts” (business or commerce, social sciences, humanities, education, creative arts, design or architecture, and law). While this may seem unsatisfactory to those looking for more precise definitions, it is the most practical way to address the research question.

Despite much reasoned debate, empirical data on cultural divisions between science and arts are scarce. Studies have examined differences in how science students and arts students think, but results have been contradictory, with some studies showing a difference (e.g. Lake et al. 2015) while others have reported no difference (Andreasen and Ramchandran 2012; Williamson 2013). Some authors have argued that interdisciplinary research collaboration is on the rise, but even these observers note that “interdisciplinary” in these cases tends to refer to disciplines that are closer together than most science-based research is from arts-based research (e.g. Porter 2009; Rafols et al. 2010). Much has been written about arts disciplines’ adoption of scientific principles to make arts research appear more rigorous, in what has been pejoratively called “scientism” (Sorell 1991)—but again, there is little empirical data on how these attitudes affect or contribute to cultural divisions in research.

Even less is known about how cultural barriers in general, including divisions between arts and science disciplines, might impede interdisciplinary research collaboration (Newman 2024b). In one smaller study of European interdisciplinary researchers, a majority of survey respondents felt that “scientists and artists belong to two different cultures” and “scientists and artists lack mutual understanding” (Vicente and Lucas 2022), but another survey-based study of researchers at dedicated interdisciplinary research institutes at three universities in the United States found “substantial long-distance interdisciplinary collaboration across the natural sciences and social sciences/humanities boundaries” (Bolger 2021). Andersen et al. (2023: 325–326) discuss some perceived barriers to interdisciplinary research but they are mainly material rather than cultural (e.g. lack of funding for interdisciplinary research, fewer jobs available for interdisciplinary researchers). While the idea that arts research and science research are culturally distinct persists (Dirks 2022), more empirical evidence would be required to substantiate these claims.

Methodology

Data for this study come from two instruments: an electronic survey of research academics, consisting of 16 closed-selection questions and one open response question, and a set of in-depth, semi-structured interviews with university leaders.

Given the sparsity of empirical research on this topic, it was not possible to access an existing set of survey or interview questions which could be adapted to the present study. Instead, I developed a new survey instrument and interview questionnaire based on an extensive literature review that I had previously conducted on this topic (Newman 202b). Survey questions were designed to address issues of cultural barriers to interdisciplinary research collaboration, with an emphasis on divisions between science and arts. Interview questions focused on the current state of affairs of interdisciplinary collaboration at Monash University, especially support for collaboration and any perceived barriers.

Potential survey respondents were identified by manually scanning all Australian campus-based department, school, faculty, institute, and research center web pages on the Monash University website. People with a standard academic appointment (e.g. “research fellow”, “associate professor”) were included, but those with a professional staff appointment (e.g. “research officer”, “lab manager”) were excluded. PhD students without an academic appointment, emeritus staff, and those with adjunct appointments were also excluded.Footnote 1In total, an invitation to participate in the survey was sent via e-mail to 2783 recipients in November 2021, with multiple follow-ups and reminders sent out to the same list of recipients in November and December 2021 and February and March 2022 (reminders were not sent in January, as this is a period of summer holiday shutdown in Australia). Invitations were sent from a Monash University e-mail address and included links to the survey, an explanatory note describing the project, and contact information for the project researchers. 430 completed survey responses were returned, for a net completion rate of 16%.

A list of 20 potential interview subjects was developed from the university’s senior leadership roster. This list was intended to capture all university leaders with senior decision-making responsibility in areas of research and impact, namely: the Vice-Chancellor (equivalent to President of the University), the Provost, the Deputy Vice-Chancellor (Research), the Pro Vice-Chancellor (Research), the Deputy Vice-Chancellor (Enterprise), the Pro Vice-Chancellor (Enterprise), the Director of the Monash Research Office, the Director of the university’s Enterprise division, the Director of the university’s government liaison office, the Director of Research Strategy, and the Deans of the university’s 10 faculties. All 20 candidates were invited to participate. 14 individuals agreed to sit for an interview, all of which were conducted between March and July 2021. Of the remaining six who were approached, three did not respond to the invitation, two declined to be interviewed, and one agreed to an interview but was eventually not able to find a workable time in their schedule to meet. Interviewees included the Vice-Chancellor, the Provost, one Deputy Vice-Chancellor, one Pro Vice-Chancellor, three Directors, and seven faculty Deans.

Interviews averaged around 45 min each. Seven interviews were conducted in person, and seven interviews were conducted remotely by video conferencing software. For 13 interviews, an audio recording was made and later transcribed to text. One interviewee declined to be recorded, so hand-written notes were prepared instead. All interviewees were promised confidentiality, in that no comments or responses would be attributed to them without their explicit consent. All interviewees were asked to sign and return a consent to participate form, which included an explanatory statement about the project and methods taken to preserve their confidentiality, and the project was approved for research ethics by the Monash University Human Research Ethics Committee (project ID 27378). Due to the small pool of interviewees, direct quotes are reported here without any specific personal characteristics, as that might identify the respondent.

Interview transcripts (and in the case of the single non-recorded interview, notes) were reviewed using basic thematic content analysis, in which interview data are analyzed and assembled like puzzle pieces to generate a bigger picture (Liamputtong 2020: 330–333). Responses were organized by coding them into a set of 11 inductively determined themes, which are presented in Table 1 (Schreier 2019).Footnote 2 Responses were also coded by direction (agreement/disagreement) to determine levels of concordance among interviewees on the subjects discussed in the interviews (Gournelos et al. 2019: 82–90).

Table 1 Themes emerging from interviews relating to interdisciplinarity.
Full size table

The method used in this study is not without limitations. As with all survey research, there is a risk of self-selection bias – that survey responses only really represent people who are willing to complete a survey on this topic, and not the broader community that the project attempted to reach. However, the set of completed surveys shows some broader representation on a variety of personal and professional characteristics, including age, gender, and career stage, which may diminish the effects of self-selection. In addition, the disciplinary backgrounds of respondents closely reflects, by proportion, the actual disciplinary backgrounds of the people on the initial survey invitation list. These characteristics are shown in Table 2.

Table 2 Descriptive characteristics of survey respondents.
Full size table

Results

Survey respondents reported high levels of experience with interdisciplinary collaboration, and while the survey did not seek to gauge enthusiasm, a majority of respondents reported that they expect to collaborate with interdisciplinary partners in the near future. For example, 88% of respondents reported that in the past they had worked on an interdisciplinary research project, 71% reported that they were currently working on an interdisciplinary research project, and 83% reported that they expected to begin a new interdisciplinary research project within the next 3 years. Additionally, there does not appear to be much difference in these responses between participants from science-based disciplines and arts-based disciplines, indicating that overall attitudes toward interdisciplinary collaboration are the same across science and arts. Difference of proportions tests did not yield any statistically significant differences in the responses from these two groups on this set of questions, with p-values above 0.05 in all three cases (see Table 3).

Table 3 Survey respondents’ experience with interdisciplinary research collaboration.
Full size table

However, despite this eagerness to collaborate, and despite the apparent agreement between science-based and arts-based respondents on their likelihood of collaborating with other researchers, it is clear from subsequent survey questions that respondents did not consider all forms of interdisciplinary collaboration to be equal. In particular, respondents from science-based backgrounds reported substantial cultural obstacles when asked about potential collaborations with arts-based researchers. For instance, only 47% of science-based respondents agreed or strongly agreed that the values of arts-based research align with their own values; 42% of science-based respondents agreed or strongly agreed that science research is more “rigorous” than arts research; and only 37% of science-based respondents agreed or strongly agreed that they understand the research objectives or methodologies used in arts research. Perhaps it is therefore understandable that a full 70% of science-based respondents reported that their next collaboration was much more likely to be with another science-based researcher rather than with an arts-based researcher.

These views do not seem to be reciprocated by arts-based researchers. 58% of arts-based respondents agreed or strongly agreed that the values of science-based research align with their own values, 69% of arts-based respondents agreed or strongly agreed that they understand the research objectives or methodologies used in science research, and only 20% of arts-based respondents agreed or strongly agreed that science research is more “rigorous” than arts research. Accordingly, only 26% of arts-based respondents reported that their next collaboration was much more likely to be with another arts-based researcher rather than with a science-based researcher.

These numbers are compared in Fig. 1, where it is evident that patterns of responses for arts-based researchers are dramatically different to those from science-based researchers. In short, science-based respondents were significantly more likely than arts-based researchers to believe in major differences between arts and science research cultures, and overall, science-based researchers were significantly more likely than arts-based researchers to report being reluctant to engage across the arts-science divide. Note that STEM is a common abbreviation of “science, technology, engineering, and mathematics“, and survey respondents would be expected to be familiar with this abbreviation.

Fig. 1: Culture-based perceptions of arts-science research collaboration.
figure 1

The values of law, arts, and social science researchers align with the values of my own research. The values of science, engineering, maths, and medical researchers align with the values of my own research. Science, engineering, maths, and medical research is more scientifically rigorous than research in law, arts, or social science disciplines. I understand research objectives or methodologies used in law, arts, or social science research. I understand research objectives or methodologies used in science, engineering, maths, or medical research. Regardless of your past experiences of research collaboration, is your next research collaboration more likely to be with a STEM/health researcher (e.g. science, engineering, maths, medicine) or an arts researcher (e.g. law, arts, social sciences)?.

Full size image

In light of these results, it might be reasonable to inquire about the potential confounding effects of variables such as gender, position, and PhD graduation year. For example, in reviewing the research culture of medical researchers from the Netherlands, Verbree et al. (2013) found that due to changes in the national research funding model, researchers who completed their PhD after the funding reforms spent more time on research and devoted more energy to external funding opportunities than previous cohorts did. The same study also found a “life cycle” effect in which the motivations and performance outcomes of junior researchers differed from those of senior researchers (Verbree et al. 2013: 176).

However, in this present study, although minor variations are apparent, the clear trend in survey respondents’ stated preferences for interdisciplinary research collaboration cuts across gender, generational cohort, and career stage—suggesting that a reluctance to collaborate on interdisciplinary research with arts-based researchers is something that is related to one’s disciplinary identity as a science-based academic, irrespective of other factors like age, gender, or career stage. This comparison is shown in Fig. 2, where it is immediately apparent that for all sub-groups of respondents, science-based researchers were more likely to report not expecting to collaborate with arts-based researchers than arts-based researchers were to say the same about science-based researchers. In Fig. 2, “early career researcher” refers to respondents who reported obtaining their PhD within 5 years prior to completing the survey.

Fig. 2: Perceptions of arts-science research collaboration by characteristics of survey respondents.
figure 2

Regardless of your past experiences of research collaboration, is your next research collaboration more likely to be with a STEM/health researcher (e.g. science, engineering, maths, medicine) or an arts researcher (e.g. law, arts, social sciences)?.

Full size image

The data would seem to suggest that these cultural divisions do interfere with researchers’ willingness to collaborate—or, more specifically, that culture-based perceptions of differences between arts research and science research present barriers to science researchers’ willingness to collaborate with arts researchers. As shown in Table 4, tests of statistical significance show statistically significant correlations at the 95% confidence level between science-based respondents’ expectations about who they would collaborate with next, and their beliefs about agreement between values of arts research and their own research, the scientific rigor of arts research, and their understanding of arts research objectives or methodologies. In other words, science-based respondents who believe that arts and science have different values, different levels of scientific rigor, or different research objectives or methodologies were more likely to also expect that their next collaboration would be with another science-based researcher. Because the Likert-scale questions being analyzed here used nominal responses, but are still ordinal in nature, the results of chi-squared tests are shown in comparison with tests based on Kendall’s Tau for ordinal data as well as the Kruskal–Wallis test, which is the non-parametric equivalent of ANOVA. All three tests showed a statistically significant relationship across all three questions, with p-values below 0.05.

Table 4 Correlation between culture-based perceptions of arts research and expectations about interdisciplinary collaboration.
Full size table

However, survey respondents were, on the whole, open to strategies for improving interdisciplinary collaboration. In response to the question, “To what extent would any of the following make you more inclined to collaborate with researchers outside your discipline?” 85% of survey participants responded that “more cross-faculty networking opportunities” would make them somewhat more inclined or much more inclined to collaborate outside their discipline—a result that puts networking near the top of the list of popular interventions, below grant funding but above workload points (see Table 5). In the open comments to the survey, many respondents commented that they had little knowledge of or exposure to research from outside their discipline and would like the university to organize more interdisciplinary networking activities. Since not all respondents provided comments, these responses should not be taken as numerically representative; nonetheless, they do provide some insight into how some respondents see networking as a potential method for boosting interdisciplinary collaboration. For example, according to one respondent, “Navigating who in the university is aligned to your research interests is challenging”. And according to another, “It is difficult to meet people in other faculties to start up conversations about what you need to do and how to do it. Some sort of central networking scheme could help this”. According to a third, “More cross-faculty meet and greet opportunities would help immensely. At this point for me personally the problem is I don’t know who the right people in my area are in other faculties”.

Table 5 Survey respondents’ preferences for improving interdisciplinary collaboration.
Full size table

Like survey respondents, interviewees were similarly positive about interdisciplinary collaboration. Although one interviewee also advised that interdisciplinary research should not occur at the expense of “core” disciplinary research, all 14 interviewees spoke favorably about interdisciplinary research collaboration in general and said they would support increasing efforts to enable it at Monash University. The most common reason among interviewees for supporting the expansion of interdisciplinary collaboration was the belief that interdisciplinarity leads to improvements in the real-world impact of university research.

Some of the interviewees seemed to be aware that tribal identities could be apparent among research disciplines—or even, that these identities might be inevitable. The common sentiments were that tribalism is a by-product of organizing the university into faculties, that it is difficult to eliminate, and that it could even be viewed as a good thing: strong disciplinary identities give researchers a sense of belonging, or community. According to one interviewee, “Every organization is tribal. People are tribal.” And according to another, “…you don’t want to deprive people of tribes. People belong based on tribes. You want people to belong.”

More generally, interviewees agreed that cultural barriers between any two disciplines can exist, regardless of whether they are arts-based or science-based. The most frequently mentioned problems related to traditional barriers such as academic culture, language, and communication between disciplines. One interviewee cited the different norms of authorship on academic publications between various disciplines as posing a barrier to collaboration, and suggested that some form of “intercultural training” – for researchers looking to publish with people from other disciplines – might be useful. According to another interviewee, communication and mutual understanding are vital for successful interdisciplinary collaboration:

…you need to bring the lens of different sorts of expertise together… You might not ask X [researcher] to be an expert in y [discipline], but you need X [researcher] and Y [researcher] to talk to one another and to discuss how they understand what they understand.

Nonetheless, not a single interviewee expressed concern that there was any kind of cultural divide between arts-based researchers and science-based researchers at Monash, or that such a divide might be a barrier to interdisciplinary collaboration. When asked directly about this issue, one interviewee remarked:

I don’t think there’s a tension between arts and science. I think there’s a historical view that they are polar opposites, and therefore they are unlikely to have things in common. I don’t think that is how it is, at Monash, actually – I think there’s a great deal of empathy between arts and science.

And according to another interviewee, “I don’t think that’s a major issue… When you’ve got excellent researchers in a room, people are usually pretty respectful of one another.” A third interviewee put it simply, “I haven’t seen it. I don’t think it’s particularly worrying.”

Even so, another interviewee ended their thoughts on this subject with a word of caution: “You don’t want tribes to be the source of discrimination, or failure to cooperate or collaborate.”

In short, university leaders interviewed for this study supported the increase of interdisciplinary research collaboration, but while they recognized that tribal identities among research cultures were possible or even likely, they seemed to be unaware that cultural divisions specifically between science-based researchers and arts-based researchers might exist at Monash University and that these divisions pose a significant barrier to collaboration.

When asked about possible strategies for improving interdisciplinary collaboration at Monash, several interviewees also spoke positively about networking. Some made reference to past networking events at the university that they thought had been successful in promoting interdisciplinary outreach. But by far the most common sentiment among interviewees on the subject of networking was the use of “virtual institutes”, where academics based in traditional research disciplines are also given an affiliation with an independent research center that has a limited administrative presence on campus. These virtual institutes act as a kind of clearing house for research: organizing seminars, running research and professional development programs, and circulating the latest research of its members. Virtual institutes have a very small number of permanent staff and little administrative authority, and also very little is required on the part of the academic to stay affiliated with the institute. By their nature, virtual institutes are interdisciplinary and span across wide sectors of the university’s academic community.

Discussion

As mentioned earlier, this study seeks to address three main research questions. First, is there a demonstrable cultural divide between science-based and arts-based academic research disciplines? The answer would seem to be that there is—or at least, that such a divide existed at Monash University in 2021–2022. Survey respondents from science-based research disciplines were less likely to report that the values of arts research align with their own values than arts-based researchers were to report the same about the values used in science research. Likewise, science-based researchers were also substantially more likely than arts-based researchers to report that they thought that science research is more “rigorous” than arts research. And lastly, science-based respondents were less likely to report that they understand research objectives or methodologies used in arts research than arts-based researchers were to say the same about science research. Not only do these cultural differences divide clearly along the arts-science boundary, but the asymmetrical nature of the results makes the divide ever more stark. A notable proportion of science-based respondents apparently saw major differences between science research and arts research, whereas arts-based researchers did not seem to perceive such a difference.

Second, if such cultural divisions exist, do they present barriers to interdisciplinary research collaboration? In this case, again, it would seem that the answer is yes. Science-based survey respondents who reported that they did not think the values of arts research aligned with their own values, or who reported not understanding the methodologies or objectives of arts research, or who thought that arts research was not as rigorous as science research, were also statistically less likely to say their next research collaboration would be with an arts researcher.

These results suggest some significant implications for the way that cultural barriers to interdisciplinary collaboration are understood. Survey respondents reported differences in values and methodologies between disciplines—or, at least, between science-based disciplines and arts-based disciplines—which lends some empirical credence to the theoretical and anecdotal literature base that has historically made these claims (e.g. Brister 2016; Lattuca 2001). Furthermore, the fact that differences were specifically visible between arts and science would seem to vindicate the “two cultures” perspective that has long dominated debate on interdisciplinary collaboration (Gardner 2013; Lach 2014; Robinson et al. 2016; Snow 1959).

However, the relationship observed here was not reciprocal. In this study, science-based researchers expressed, on average, skepticism about the value of arts-based research and a corresponding reluctance to collaborate with arts-based researchers, but responses from arts-based researchers did not display the same trends, with arts-based researchers expressing a general willingness to collaborate with science-based researchers. This outcome suggests that a re-think of the two cultures perspective might be in order. Importantly, in this study, the sentiment that science research is more rigorous than arts research was twice as common among science researchers as it was among arts researchers—suggesting that science and arts may not be divided into two co-equal cultures, but rather that more nuanced dynamics exist that would need to be investigated further. Arts-based researchers may want to collaborate with science-based researchers because they believe that such a collaboration would lend some “scientific” legitimacy to their arts-based work, which would support the view of the “scientism” literature that argues that arts-based researchers suffer from a kind of inferiority complex when compared with science-based colleagues (e.g. Sorell 1991). These asymmetrical results were unanticipated in this study and the survey instrument was not designed to capture this detail.

The third research question asked: if cultural divisions between science and arts can be shown to hinder interdisciplinary collaboration, what are institutional leaders doing or planning to do to mitigate the effects? In this case study, institutional leaders did not seem to be aware of any arts-science cultural divisions—which may effectively be a barrier to interdisciplinary collaboration on its own, as previous studies have shown that strategies to encourage interdisciplinary collaboration can require institutional support in order to be successful (e.g. Gibson et al. 2019; Siedlok et al. 2015). It may therefore make more sense to ask instead: if interdisciplinary collaboration is desired, and if cultural divisions between science and arts can be shown to hinder interdisciplinary collaboration, what should institutional leaders do to mitigate the effects? The cultural divide between science and arts observed in this study appears to be pervasive, cutting across a variety of demographic characteristics such as gender, age, and career stage. It would seem unlikely, therefore, that small-scale targeted interventions could be effective. Instead, a larger, perhaps even long-term, university-wide strategy would be required.

Perhaps the simplest way to bring divergent cultures together is through exposure. While the dynamics are somewhat complex, research has demonstrated that exposure to different cultures can improve empathy, tolerance, and understanding (Bruneau and Saxe 2012; Hoter and Shapira 2022). In addition, studies in cognitive science have shown that simple exposure to new ideas can increase creative thinking (Chen and Althuizen 2022; Fink et al. 2012), which is a key desired element of interdisciplinary research collaboration. By this reasoning, strategies that promote exposure between arts researchers and science researchers may help to bring these different research cultures closer together. In this study, increased networking among researchers was cited by both survey respondents and interviewees as a desirable and effective way to improve interdisciplinary collaboration. A potential pathway to improve interdisciplinary research collaboration, especially between arts and science, may be to get these researchers more familiar with each others’ work through university-led networking activities and ongoing organized networking structures such as the “virtual institutes” favored by this study’s interview participants.

Conclusion and future research

In this study, a survey of academic staff and interviews with senior university leaders have revealed a cultural division between science-based and arts-based researchers at Monash University. While support for interdisciplinary research collaboration is high among Monash’s leaders and academics, science-based researchers have reservations about arts-based research and also appear to be reluctant to collaborate with arts-based researchers. The reverse was not observed; arts-based researchers did not seem to have any qualms about science research or about collaborating with science-based researchers. This dynamic presents barriers to interdisciplinary research collaboration at Monash University, and will therefore make it harder for the university to implement its stated strategy of increasing the quantity and diversity of interdisciplinary research. From a more theoretical perspective, the results of this study suggest that a rethinking of the two cultures debate might be in order: cultural divisions between arts and science may actually exist, but the tensions may only run one way.

As discussed earlier, there is nothing particularly special about Monash as it compares with other high-performing research-focused institutions in Australia and in other countries around the world. Moreover, given the high levels of mobility of the academic workforce in Australia and other countries (Heffernan and Heffernan 2019), it is likely that many survey respondents had recently worked at other universities and equally likely that many would soon move on as well, bringing or taking their cultural views on research disciplines with them. While it is, of course, difficult to generalize from a single case study, the results presented here at the very least suggest that this issue may be evident at other similar institutions in Australia and in other countries.

However, of course, the data presented here do come from a single case study, and further research would be necessary to confirm whether tribal identities or cultural differences might impede interdisciplinary collaboration at other research institutions as well. It is possible that pressures related to research performance are higher at research-focused institutions, and that these cultural divisions may not be as prominent at smaller universities and research organizations, where pressure to perform is lower and the academic community is smaller and closer together, or where a greater emphasis may be placed on teaching.

These results suggest some avenues for further research. First, wider studies would need to be conducted in order to establish any broader generalizability of the results presented here. Future survey-based studies should invite participants from multiple universities, and ensure that different kinds of universities (research-focused as well as comprehensive, large institutions as well as small ones) and institutions from different geographical locations (urban as well as rural, and institutions from multiple jurisdictions) are represented. This will help determine whether the results found here are evident outside of Australia and what differences might be related to size or focus of various research institutions.

Secondly, why does the arts-science cultural divide indicated in this study only run one way? Is it possible that arts researchers, despite claiming fairly emphatically that science is no more rigorous than arts, seek to collaborate with scientists to lend legitimacy to their research, perhaps with an eye toward better success in obtaining grant funding? In depth interviews with arts researchers and science researchers might uncover some deeper nuance into this dynamic.

And finally, the dimension of personal experience was not explored in depth in this study. It is possible that some researchers may have had negative (or positive) experiences with interdisciplinary collaboration in the past that might influence their eagerness to collaborate in the future, especially across the arts-science divide. Future studies might ask more specific questions about respondents’ personal experiences with collaboration, including the reasons why they may be reluctant to collaborate with researchers from particular research backgrounds. This could be followed up with in-depth interview data to explore this question more fully.

In this study, university leaders did not seem to be aware of any cultural divisions between science-based and arts-based researchers at Monash University. In 2022, the results of this study were presented to the university’s executive, along with a comprehensive report. Around the same time, the university developed and released a new strategic research plan that puts more emphasis on interdisciplinarity, collaboration, and research impact. Like other universities, Monash is responsive to the modern environment in which public universities are increasingly required by their government funders to justify their utility to society. While discipline-based research is likely to remain an important cornerstone in the advancement of knowledge, institutions that are able to cultivate an appetite for interdisciplinary collaboration across their academic communities may be better placed to demonstrate the utility and impact of their research.