the periphery focuses on empirical case studies about topics relevant to the author’s home community, drawing on these foundations.
In most cases, centre/periphery concepts refer to places where certain power relations are played out: the distinction is tied to nations or actors who occupy a physical place or space. From those positions, control is exer
cised over the knowing and doing of other actors, positioned elsewhere (ad
ditional to those referred to already; see especially Strassoldo 1980). To con
clude this section, I see three problems with these types of concepts. Firstly, they have to admit an enormous number of exceptions. The Kenyan writer that is nominated for the Nobel Prize for literature, the Philippine start
up that a whole line of business goes crazy about, and so on. A sound analytical concept, embedded in a theoretical framework, must somehow address the possibility of outstanding events like that. Secondly, if, for some reason, a country or a location is labelled as peripheral, it is stigmat
ised with everything it physically ‘contains’: people, businesses, culture, et cetera, while individual status and selfobservations might lead to very different diagnoses with regard to a centre/periphery position. Thirdly, in the spatial conceptualisations reviewed, this stigmatisation does not entail much leeway for change. I agree that the economic, political, medical, and educational conditions in many countries are calamitous for the ma
jority of the populations, and people have every reason to feel ‘damned’, but I claim that a sound theory must be capable of incorporating possible change, not least to identify points of potential intervention. This is the motivation behind my research.
3.3 Centre/Periphery as Inner Differentiation
to an unbearable level, this is where peripheries are produced. They appear to be spatial because of accumulations of attributions of peripheral status in certain world regions, so these regions are designated as peripheral.
As I discussed above, an allencompassing spatial view on the periphery seems to be the standard in the literature. Von Gizycki’s definition of the peripheries deviates from this standard (1973): the origin of work highly acknowledged by a certain community defines a scientific centre, either the worldwide scientific community, or, most often, communities on dis
ciplinary or regional levels. Vanpaemel (2012) (who supports the standard definition), criticises that this leaves the periphery in a passive mode, but I see this as a typical misunderstanding, derived from observing communic
ation not as communication, but as attributed to certain people. I am not claiming that von Gizycki is suggesting a preadaptive social systems theory concept, but in my reading of his text, there is a foreshadowing of a sim
ilar understanding of communication. This also becomes apparent when he cites Pasteur with ‘Scientists have a country; science has none’. Von Gizycki never says that only researchers working in regions with central status have the communicative power to add to the assignment of centrality to certain places with their contributions. Peripheries actively contribute to the construction of centres.
Blurred and flexible spatial boundaries between centres and peripheries naturally emerge from contributions to scholarly communication. Con
sequently, a definition of the distinction needs to account for them. Ad
ditionally, a communicative definition emphasises the significance of the periphery to the centre: theories and methods developed by the centre, in order to confirm its central status, must be acknowledged by the periphery (cf. Bhambra 2014a, p. 81). The centre is more dependent on the periphery than the other way around, because it relies on highly inflexible institutions, where fears of betraying their trajectory are always present. In contrast,
‘[t]he periphery can change and reorient itself ’ (von Gizycki 1973). As Vanpaemel (2012) concludes, ‘the perspective is shifted from a competit
ive model [describing how the periphery is emulating the centre] towards a complementarity model’, in which the periphery is just as significant as the centre.
Luhmann suggests a similar explanation for why the centre/periphery differentiation is useful in society and therefore has not been overcome so far, although it reduces opportunities of action for large parts of soci
ety: maintaining a centre that works on the forefront of what is perceived as new is risky—it costs a lot of resources. If something goes seriously wrong in the centre, the whole system will remain functional only because of the periphery (Luhmann 2005, p. 250). While the periphery could very well continue to exist on its own terms, the centre depends on the system
stabilising effect of the periphery (Luhmann 1992b, pp. 355 sqq.). The centre therefore also depends much more on this form of differentiation, and this is what creates conflict: the periphery is exploited as some kind of backup system. Its capacity to serve the main task of the research sys
tem, to provide society with new knowledge, is obscured by this backup function. Furthermore, the overlay of the communicative with the spatial concept stabilises the references to places as peripheral, so central commu
nication can reassure itself of itself as central.
Contrarily, Keim (2008, pp. 53 sqq.) suggests a somewhat communicat
ive model of centre and periphery that focuses on the content of the schol
arly communication—of sociology, for that matter—and includes spatial correlations only as precondition and effect, in terms of infrastructural de
velopment. At the core of this definition is the reliance of peripheral and dependent sociology on cooperation with central sociology in order to con
tribute to it. Peripheral sociology lacks autonomy. Further, it is charac
terised by a lower level of abstraction, and studies its direct environment only,⁷⁵ without being well embedded in and accepted by it—not least be
cause of a merit system that places greater weight on foreign appreciation (Mkandawire 1989). This multilevelled definition might capture the situ
ation in places that are perceived of as peripheral, and it does provide some descriptive flexibility, but it offers only a diagnosis, and no potential for
75 Keim 2008, pp. 132 sqq., confirms the peripheral status of African social sciences by analysing selfreports by research institutions from the now ceased UNEScO database DARE, in 2003: while 60% of German or French institutions in the sample focus on con
tinents beyond Europe, only 6% of African institutions study populations outside their home continent.
change of typical perceptions, unless they are simply cleared. In the com
petition for centrality, peripherality will always be the losing part that is left behind (Keim 2008, pp. 110 sq.). Compared to that, social systems theory offers a concept that sees important systemwide functions on the peripheral side of the distinction, which are attributed not to researchers or organisations, but to communicative contributions.
Without arguing functionally, Vanpaemel (2012) criticises any claim of a ‘unidirectional spread of Western Science’ as modelled by Basalla (1967), and supports this with the historical studies of science by Raj (2007) and Gavroglu, Patiniotis, et al. (2008). Rather than a ‘simple emanation from a preexisting centre’, ‘Western Science’ was ‘the result of complex processes of conflict, acculturation, and appropriation’ (Vanpaemel 2012). While I do not doubt the reliability and value of studies which reconstruct the cooperative conditions of knowledge production and reception with a ‘cir
culatory perspective’ (e. g. Raj 2013), I suggest acknowledging the differ
ence between contexts of production and the actual dissemination of the new knowledge via publication, in which the cooperation with indigenous scholars is obscured. Their role is not acknowledged in the same way as the role of European scholars, their names are omitted. Consequently, for the research system, indigenous scholars have no addresses for attribution; they cannot be referred to unless a historian of science creates these addresses posthumously. This historical type of exploitation for research purposes is what then developed into the ‘academic division of labour’, which would normally, or, depending on the specific role, occasionally, acknowledge the contribution of the indigenous scholar with a subordinated authorship, and therefore create a subordinated address (for a number of examples, see Esseh 2011, p. 92).
A communicative concept of centre/periphery can also be applied when single social systems are observed, without stigmatising places in an all
embracing way. When employing the concept as an analytical tool, one is forced to observe communication very closely: when and for what reason is a peripheral status assigned and in which communicative context? In order to better understand the implications of the concept of centre/periphery
in social systems theory, I will draw on its role in Luhmann’s historical considerations. Two simple examples, from the contexts of rural economy and spirituality, shall help to prepare the understanding of the complex case of current scholarly communication.
While Luhmann describes an evolution of ‘Western’ society that culmin
ates in proclaiming world society, he still does not buy into a universal line of human social development (Epochenbildung) (1997, pp. 609 sqq.).
Different forms of differentiation that structure society/communities over
lay one another, changing in dominance over time. When people started farming, their society was typically organised in households and/or com
munities that exchanged and shared goods, knowledge and customs with other communities within reach. Only when a centre for this exchange was established, like a market or a town, did the primary type of differ
entiation change from segmentation to a centre/periphery differentiation.
Then, the periphery would extend as far as the centre exercises attraction to perform certain activities mainly there, and according to specific cus
toms established there. Different expectations of how the communicative exchange works, in regard to, for example, trading or spiritual rituals, are typically bound to certain types of differentiation; for instance, personal ac
quaintance might no longer be expected from all participants in a spiritual ritual of the centre.
The physical location identified as the centre is important for commu
nication as a point of reference, and conditions such as whether there are, e. g., shady trees in the marketplace or not might have a huge impact on how the market is organised and how people interact. However, the trees only become important for an analysis of communication if this impact can be detected in the communication, implicitly or explicitly. Staying with these examples, peripheral would be spiritual activities happening in a community which is aware of the rituals performed in the spiritual centre, but which continues its own spiritual activities—deliberately or not. Peri
pheral would also be any trading happening in the area of influence of the central market, but which, for instance, has established deviating ‘market rules’ and remains separated from the central market’s action—deliberately
or not. As long as the central procedures stick out, peripheral activities re
inforce the central status by contrasting it, and thereby highlighting some of the centre’s features.
How can communication, loaded with attributes such as an affiliation that carries a lot of peripheral meaning, possibly move to the centre? Das
gupta’s (2009) discussion of ‘progress in science’ in peripheryPeripheries can help to find an answer:⁷⁶ progress, meaning opportunities for own re
search programmes, presumes that peripheral researchers engage creatively with an ‘unsolved issue within metropolitan science’. This resonates with a communicative concept of centre/periphery. It is quite unlikely—but not impossible—that this paratext can coexist with scholarly claims that end up in the system’s centre. To begin with, the references of such contribu
tions to claims that are already in the centre must be extraordinarily strong so that the scholarly ‘value’ is recognised.
Centre/periphery is not the only form of internal differentiation of the research system.⁷⁷ For instance, it is also functionally differentiated intern
ally, in disciplines, and again in research areas or fields. Internal functional differentiation creates an internal environment for each subsystem, that therefore helps to reduce complexity, and to mediate the environment ex
ternal to the research system: relations between subsystems can be of help when there is outside pressure (Stichweh 1988, p. 86). These subsystems are differentiated internally even further: for each research area, its rep
resentatives will more or less agree on where they locate the world centre or centres of the research area. Instead of referring to certain countries or regions, it is likely that specific institutions are pointed out as centres of these subsystems. Often, disciplinary centres concentrate in one location because they share facilities (Luhmann 1997, p. 673), such as public at
tention, administration, representative lecture halls, and libraries (also see Altbach 1981). As in any other region of the world, in Southeast Africa there are locations where these facilities accumulate, and these are perceived
76 Dasgupta does not refer to Galtung, but while the meaning that Galtung gave to
‘peripheryPeriphery’ is what he works with, he never uses this label.
77 For a comprehensive list, see Stichweh 1988; for the theoretical roots, see Luhmann 1997, pp. 760 sq.
of as regional centres of research (see Section 4.2).⁷⁸ Where the commu
nication places the centre always depends on the topic, on participation, on the composition of further self and otherreferences, et cetera, within a row of communicative sequences; and to be remembered as centre, the denotation has to be repeated very often.
Derived from this concentration, it becomes increasingly difficult to es
tablish centres elsewhere. However, when facilities become seriously out
dated, libraries are run down, reputed researchers cannot be attracted, and therefore communication does not coincide with that place any longer, a centre status can be lost. Normally, the differentiation reinforces itself, and even if this does not have any function for science, it simply comes with the package of this form of differentiation (Luhmann 2005, p. 79).
Furthermore, that package also contains the selfdescription of the centre as superior, upon ignorance of the periphery. Everything that does not belong to the selfperception, or that creates problems which cannot be solved, is externalised (Luhmann 1997, pp. 854 sq.).
The current need for distinguished centres, longterm focal points for a discipline, can be questioned. One rather hardtorealise option is ‘scratch
ing off the whole thing’ and rejecting the concept of one research system that developed with strong references to Europe all along the way. For in
stance, this is demanded in the widely discussed video of a discussion at the University of Cape Town.⁷⁹ In this line of thinking, it is preferred not to build on the foundations of the often racist, exploitive and discriminative history of research. The idea is to position alternative systems based on ‘in
digenous knowledge production’ and reception opposed to the established research system. Alternatively, it is demanded that the research system
78 Those accumulations, including human and nonhuman ‘actants’ are constitutive for what Latour 1988, pp. 179 sqq. denotes as ‘centres of calculation’.
79 Science must fall?, YouTube, published on 13 Oct. 2016 by UcT Scientist, https:
//www.youtube.com/watch?v=C9SiRNibD14, visited on 29 June 2020. According to Roy 2018 ‘the phrase also runs the risk of being used by religious fundamentalists and cynical politicians in their arguments against established scientific theories such as climate change’.
Furthermore, Roy continues, the call for full rejection also does injustice to critical and progressive scientists from postcolonial countries.
should accommodate a variety of ways of doing research (see e. g. Alvares and Faruqi 2012; Alvares 2014; Anyidoho 2008; Bowker 2010; Sitas 2004;
Smith 2012; Wilson 2008). ‘Indigenous knowledge production’ and recep
tion potentially provides new knowledge not only to a certain community, but to all of society.
Multiple and mobile centres of the research system, while certainly re
quiring at least a wider distribution of resources, are not putting the whole system at risk. Since research institutions have been created in almost every country of the world now, risk can be distributed more evenly, to stop the exploitation of peripheral research as verifiers of research results from the centres and data collectors in collaborative projects (see e. g. Holm, Jar
rick et al. 2015, pp. 114 sqq.). According to dependency theory, even if collaborations are set up on equal intellectual footing, contexts where schol
arly centres accumulate always benefit most from the projects, because of their ‘advanced capacity to disseminate, absorb and act upon knowledge produced’ (Boshoff 2009, with reference to Nagtegaal and de Bruin 1994).
Therefore, research projects need to take responsibility in order to com
pensate for those effects. ‘Additive’ approaches are insufficient to tackle the social injustice that is reproduced by the system, regardless of whether they concern data collection in remote places, far from established centres, or the implementation of selected ‘indigenous’ concepts or methods into the research system (cf. Bhambra 2014a, p. 81).
The world economy is described as being increasingly decentred (see e. g.
Lash and Urry 1994, p. 4). ‘Human life, it is also argued, is increasingly deterritorialized’ (Hannerz 2015, p. 311). Even if that was true (only) for those who can afford mobility, and who therefore are likely to claim some
thing like that, it leaves society with open questions about how to acknow
ledge and tackle injustices. At least in the case of research communication, this can be done by analysing how the differentiation of centre and peri
phery is created in communication.
Most deeper analyses of the locality of academic knowledge focus on historical examples from the natural sciences (e. g. Livingstone 2003; Del
bourgo and Dew 2008), and therefore mostly on implicit references in re
search communication to European territory. The reason for that might be
that in the SSH, with the exception of the field of history (see e. g. Chakra
barty 1992; 2008), universality claims are less often made. However, they are abundant in theories and methods (see e. g. Mignolo 2000; Sitas 2004;
Connell 2007; Smith 2012). Yet the locality of knowledge must not neces
sarily lead to an internal differentiation of centre and periphery that is per
ceived of as racist and neocolonial. This perception cannot be explained on the basis of historic trenches alone, but is also due in part to current stabilising mechanisms built into the research system, some of which I will describe below.
The scholarly communication system is observing itself in many forms;
on the one hand, with the help of reflection theories (epistemology, sci
ence studies), and on the other, by quantifying selfobservation (citation databases, rankings, institutional audits, et cetera whose sound methodo
logy has to be accepted—with good right or not). In any ranking, the idea of comparability is implied, while especially for university rankings, this is, as Connell (2019, p. 85) notes, a ‘fantasy’, in which ‘[s]erving distinctive local needs counts for nothing’.
The visibility of a cited publication is multiplied by its citations, and this effect is reinforced by selfobservation techniques, resulting in a Matthew effect.⁸⁰ Many citations increase the reputation of a certain author, whose future work is automatically more visible. Institutions with a long list of reputable affiliated scholars rank highly, and therefore are specifically at
tractive places at which to conduct research, and so on. Because of these structuring effects of the research system’s environment, entire world re
gions appear peripheral, because references in the ‘centre literature’ rarely point to literature written or published there. The effect is a neglect of whole regions in global selfobservation exercises. Such a gap can then become a reference of followup communication.
This selective process of not recognising everything that has been pub
lished tackles the problem of volume that the system has to address some
80 ‘The Matthew effect may serve to heighten the visibility of contributions to science by scientists of acknowledged standing and to reduce the visibility of contributions by authors who are less well known’; see Merton 1968.
how: it usually is sufficient to include ‘centre literature’ in a literature re
view; as WOS founder Garfield says, ‘a few hundred core journals would satisfy the needs of most readers’ (Garfield 1997). Stichweh talks about a
‘legitimate constriction of the view’⁸¹ that works as long as all crucial re
search innovations that might happen in the periphery are absorbed into the centre. As Vessuri, Guédon, et al. (2014) make clear, work that gained some visibility, e. g. by being retrievable via GS, or even published in a
‘Global North’ journal, is not automatically positioned in the centre (see OndariOkemwa 2007): ‘visibility alone is not enough. Effective presence requires being in such a state of visibility that anyone neglecting it will be faulted for carelessness, incompetence or ignorance’. At the same time, every centre of scholarly communication in history brought a restriction of the acceptable and possible with it, a homogenisation (Stichweh 1988, p. 92). With the advanced technology available today, homogenisation and Matthew effects could probably be minimised.
The large output characteristic for centralised places becomes noticeable because it clusters around certain ideas and topics, and timely participation via selected media is key to gaining attention. Uk and US scholars have a head start for various historical, linguistic, and infrastructural reasons, but their contributions also do not always succeed in the run for global reach.
Peripheral communication is clearly linked to central communication, yet sometimes delayed. It might thus miss the hot topics, and it uses media, including language, that can make it hard to overcome the improbability of successful communication. Paratexts of prestige also help to overcome this barrier (also see Schmidt 2016c). To repeat that point, peripheral com
munication can only live up to the expectation of functionally differenti
ated communication in world society because it makes global reach happen.
This global reach is a requirement of the function system in world society, and not a requirement of local knowledge communities. In other words, if the peripheral scholarly communication referentially attached to a large
81 ‘Legitime Sichtverengung’ in Stichweh 1988, p. 91. Similarly, von Gizycki 1973 states that ‘being at the centre, their members may no longer feel the need to inform themselves about what is going on in certain peripheral countries where serious scientific exertions might be taking place’.