4 Development Cooperation as a
Field of Transnational Learning1
Kay E. Ehlers and Stephan Wolff
LEARNING IN INSTITUTIONAL SECTORS
The perpetuation of interorganisational learning depends on the shared learning interests of cooperation partners on one hand, and on interorganisational learning barriers on the other. Learning in and from complex institutional arrangements cannot be regarded—or only to a very limited extent—as the consequence of the efforts and successes of the participating actors. The systematic reference to the deep structures and process rules of the organisations involved is therefore essential for the theoretical reconstruction and the empirical research of learning. The conditions and possibilities of organisational and interorganisational learning (or lack of learning) can only be revealed against the background of those institutional specifics (cf. Ingram 2002). In this chapter, we will attempt to trace this issue using the example of development cooperation (DC). Learning in this large sector is (or should be) an indispensable condition closely linked to both the sector's transnationalism as a set of organisational arrangements and effective social support in developing countries: Which types of organisational set-ups perform better than others? Which lessons have to be learned about successful and failing strategies? However, first of all the problem to what extent organisations and individuals are motivated to learn needs to be analysed.
DC is perceived as the bridging, if not the abolition, of blatant differences of an economic, social, cultural, or spatial type. To this extent, it transcends a goods market, where the defined goals of the organisations involved are to some degree effectively and reliably fulfilled in the exchange of goods/services for money. Hardly anything exists that cannot be the object of DC. Processing decisions and actions in DC, moreover, are much more broadly defined than the repertoire of a single functional relationship (health care, education/training, economy, etc.). Further, DC is characterised by the considerable diversity of the organisations participating in it: consulting firms with profit-making intentions, non-governmental organisations, bureaucracies of the developing countries, development banks, construction companies, UN organisations, village self-help groups—all come into contact with one another.
For this reason it is difficult in DC to implement a generally binding purview of meaning beyond the customary harmonistic rhetoric. A theoretical conception of such complex interconnectedness is intended by the neo-institutional concept of societal sector (Scott and Meyer 1991: 117 ff.). A sector means here a collection of organisations in a certain social area, which make a contribution to the production of certain goods or services, or influence the central organisations that produce these goods and services. Functional and institutional aspects, rather than geographical ones, are crucial in the definition of the limits of a sector.
Learning cannot be conceived of as an accumulation of knowledge and ability accessible to all the actors in that sector. A sector is in certain ways integrated. Without question there are common points of reference, global objectives, and binding modes of language. However, there also are—unavoidable in such an extensive ensemble—organisational peculiarities of the most varying kinds. Luhmann (2000: 409) writes of the “contradictory unity of interdependence . . . while at the same time organisations have symbiotic relationships with one another.” The conditions of a sector and the interorganisational relationships linked with these conditions create a certain structural resistance against impulses towards change and limit stimuli to learning. Therefore, interorganisational relationships are certainly not fields of learning par excellence. The organisation of learning and knowledge is subject here to a selection that, to a large extent, only arbitrarily favours the declared objectives of development. Between the world of the target groups and that of the strategic decision-makers, a barrier arises that impedes learning processes with diverse organisational ‘impurities.’
The general background to interorganisational learning in development cooperation outlined in the following discussion draws upon the results of an empirical study on the sectoral manner of functioning of development aid projects (Ehlers 2011). The four medium-sized projects discussed here lasted several years, were financed by West European development banks, and took place in different West African states from the mid-1990s on. Their common objective was the improvement of access to drinking water and to basic sanitary supplies in selected regions and cities. Different actors from various organisations of the DC sector participated in the measures aimed at hygiene education, water marketing, organisation development, and the building of simple sanitary infrastructure.
These organisations included consulting firms from European industrial countries and local consulting firms, donor organisations and development banks, as well as the responsible administrations, departments, and offices of ministerial organisations or of the provincial government. These vastly diverse organisations, each of them grounded in their own contexts, and their collaboration occurring in an organisationally specific transnationalism, form the core elements and processes of project-based development cooperation. This fundamental structure has to be complemented with other, more peripheral organisations and actors. They include other donor organisations from the same sector operating in the same recipient country, non-governmental organisations with or without administrative backing, as well as ‘traditional’ structures such as tribal chiefs and clan councils. Furthermore, there are influential individuals (such as local politicians) whose organisational anchoring is unclear, but who have been attributed competence and/or importance. Finally the target group's elements are paramount and must not be forgotten according to the claims and programme. They may consist of neighbourhood communities, villages, or the beneficiaries of training programmes.
This organisational complexity is augmented by substantial differences or even antagonisms between individual persons or departments within the organisations. For example, European consulting firms also employ citizens of the recipient countries at the expert echelon, though less at the executive level. Moreover, one often confronts conflicting interests (or reciprocal deficits in observation) in the administrations of the developing countries, e.g., between government bureaucrats and those from offices in the provinces. Finally, one has to take into account evaluations, either from commissioned evaluators or from delegations from the donor organisations.
THE RIDDLE OF BLOCKED LEARNING IN DEVELOPMENT COOPERATION
The question of whether and how DC can learn from its own experience in order to improve its performance or performance ability remains open today (cf. Wright and Winter 2010). This is surprising, since in the debates of the 1980s on aid effectiveness, donor organisations as well as their partner organisations in the recipient countries were advised to pay more attention to organisational learning. The self-proclaimed transformation of the World Bank into a “Knowledge Bank” by its president during his inaugural address in 1996 lent the so-called knowledge and learning approach additional prominence, laying the foundation for today's dominant idea of development aid based on knowledge and learning (Ramalingam 2005).
Nevertheless, the assertion by Cassen et al. in 1986 still seems to apply: The question remains unanswered of whether and how international development cooperation can systematically derive from its experience the capacity for the purposeful improvement of its future performance. Typical actors such as representatives from SADEV2 self-critically admit that in the field, “despite increasingly rigorous feedback systems, development agencies continue to be criticised for their inability to incorporate past experiences. They are routinely accused of learning too little, too slowly—or learning the wrong things, from the wrong sources” (Krohwinkel-Karlsson 2007: 6). Therefore, what should be cleared up is: Who did not learn here and why not?
Those in the relevant field reflect similarly on the issue. Thiel (1998), editor-in-chief of the German journal Entwicklung und Zusammenarbeit (e+z)3 [Development and Cooperation], criticises what he regards as an obvious scandal. On the one hand, he argues, there is hardly an area of politics that is as strongly marked by continuing learning processes as development politics—because of its supposed nature and raison d'être. Evaluations (often conducted on the German side by the BMZ4) represent one of the most important resources for the scrutiny of the public, which can find out how many of the financed projects were successful. For the participants in the aid process, the results of such evaluations can help to create a better basis for future work. On the other hand, Thiel states, learning is continually hindered in practice, because the development bureaucracy in effect claims that the learning process can only take place inside the bureaucracy, and operates in such a way that the public, instead of being involved in the process, is merely informed of the result.
In other words, Thiel's assessment posits that the aid development bureaucracy learns, but it does not allow the public to participate in this learning. By withholding a series of evaluations, the BMZ consciously—or at least negligently—hinders the learning of other participants. He detects an inbred or accepted lack of transparency, the elimination of which is increasingly called for by the recipient countries. Why does such obfuscation arise? Thiel's criticism, born of long experience, can be summed up in four assumptions:
| Assumption 1: | No one wants to admit to the uncomfortable truth that single projects have only a slight effect on the development of a country. |
| Assumption 2: | Talking about projects and their results becomes too abstract and incomprehensible to the various stakeholders. |
| Assumption 3: | International aid in developing countries has bred a new parasitic clientelism, centred on intermediaries, the so-called development agents, who can only exist because they understand this jargon and take advantage of it, but who have no particular interest in the successful realisation of projects and their evaluations. |
| Assumption 4: | The government ministry in charge exerts pressure on its employees to not reveal details of their knowledge, with a view to either avoid demonstrating political weaknesses or in order to reserve for itself the prerogative of interpretation vis-à-vis other institutional colleagues. |
These assumptions imply that, with the good will of all those involved, DC could in fact be effectively designed not just as regards its objectives, but also as a learning process. They likewise signal where one should begin in order to truly make DC a learning relationship between organisations:
Observers of the political situation in development cooperation must be perplexed that complaints about the alleged learning deficiencies of DC have been lodged in identical form literally for decades, in spite of the notorious denunciations and apparently available corrective measures. Such persistent complaints indicate that when one attempts to comprehend the phenomenon, traditionally assumed causes (such as ignorance, lack of education, political calculation, or scarcity of resources) do not bring one further. Instead one has to change perspectives, i.e., re-specify the problem. This is what we would like to do, by describing and analysing the criticised learning barriers as representing both the expression and the result of a certain organisational arrangement.
STRUCTURE-GENERATED LEARNING BARRIERS
Learning in the DC sector, whether the learning of individual organisations or of the sector as a whole, is apparently a complex and subtle process, involving objectives and standpoints that are extremely different, difficult to recognise, and that are not easily amenable to form a consensus. To expect anything else would mean disregarding the existing segmentation of a functional relationship in simultaneously operative institutions, as much as disregarding a more understandable division into cooperating and competing organisations. Since learning processes (have to) take place in this framework, wanting (or expecting, or even claiming) to recognise in them a ‘pure’ or ‘strictly objective’ occurrence is not very promising. Instead, one should first of all attempt to reconstruct the organisational and institutional conditions under which learning—in a broad sense—takes place in (via/with) development aid projects.
Figure 4.1 The aid octangle (Ostrom et al. 2002).
Certain types of organisations that operate in the DC sector can be grouped into different classifications. The most prominent modelling of the pertinent interorganisational structure is the ‘aid octangle’ devised by Elinor Ostrom et al. (2002). As Figure 4.1 demonstrates, one has to expect a specifically accented configuration of institutional players that varies with each DC measure.5
Learning processes and knowledge accumulation in the DC sector can be visually rendered and in greater detail using the central sub-processes of the octangle. Three core partners are always essential for the realisation of development aid projects: the donor organisations (or development banks, aid agencies), the recipient organisations (for example, state administrations, sectoral ministries/agencies), and the consulting firms (contractors), whose participation, particularly in difficult enterprises, has hardly been considered in the literature.6 We would like to now provide a structural illustration, and then use empirical vignettes to illustrate some key mechanisms in the field of learning and knowledge that characterise the exchange between these three types of organisations.7
Learning, in the sense of transfer and appropriative transformation of information about innovations, occurs in the interaction among participating actors of the DC sector. From the perspective of observers, this should lead to the build-up of a continuously improved knowledge base and ultimately to changed patterns of action. This optimistic conception of a cumulative sectoral learning gain across time can be explained by Figure 4.1.
Figure 4.1 suggests that the knowledge base of the organisations involved (indicated by the size of the circles) increases across time, fed by a constant stream of information (while the intensity of this stream itself demonstrates an increase). The total sectoral system (DC t1, DC t2), which can be conceived of as the sum of individual development aid projects (ΣPit1, ΣPit2), is on a growth path during the course of time (cf. the ordinate indications and the arrows).
Figure 4.2 Optimistic conception of interorganisational learning.
The model conception has to be confronted with the particularities of DC practice and tested for empirical validity. We will do this with the help of case vignettes that reproduce typical interorganisational configurations where informing and learning (could) play a role: the reporting back of the consulting firm to the donor organisation on the course of the project (‘the report’), the storing of accrued information at the consulting firm (‘archive’), the direct exchange of information between all those involved (‘the on-site meeting’), and finally the learning of a semantics appropriate to a development aid project by the recipient organisation (‘the management of terms’).
THE REPORT8
Let us imagine a model conception as shown in the figure of a first key scene of a DC project that is professionally carried out: the learning relationship C >>> D, which is information typically offered to a donor organisation by a consulting firm. This relationship is manifested primarily in the periodic (for example, quarterly) report9 during the course of the project. Such reports should not only transmit information. Provided that the characteristic of triggering learning is to be ascribed to them, they should be uncomfortable and irritating, too.10
Figure 4.3 Consulting firm and donor organization.
This, however, is not the style in which such reports are customarily written, as the usual term for them used by donor organisations clearly indicates, ‘progress report.’11 This designation is not an invitation to the author to reveal his or her irritation, criticism of the task, of not knowing how to proceed, or other deviations from the apparently expected normal process of progress. Such a response would constitute for the involved organisations, D and C, an unpleasant, supplementary imposition, ultimately undermining their raison d'être. D expects the achievement of the objectives set and does not wish to present itself as the source of tensions and is not designed to process information such as “it doesn't work,” but rather to proclaim statements of the type “once more the concept proved itself.”
It is therefore necessary, particularly for individual actors who take their responsibility to their own organisation seriously, to demonstrate in these reports that deadlines have been kept, that quantitative objectives have not been fallen short of, or agreements decreed as sensible in the project conception could actually be made. It can make sense to refer to the regular course of a visit by a delegation from the donor organisation, or to an invoice that will soon be due for payment. This specific choice of topics, with its focus on standard procedures, is likely the normal case, but in no way should it degenerate into a triumphal report. Without a doubt there are conditions when this pattern is interrupted, as when the realisation of the project faces serious political resistance.12 Then learning has to take place or—when avoiding it continues to be advantageous—D has to make increased efforts in order to shield the customary internal patterns from the imposition of an irritation.
The periodical report from C to D ideally signals above all one thing: C does not challenge D with anything other than D's own expectations and claims. D accepts the report; the implied message is that D does not really want to know exactly where the fodder for irritation (from the viewpoint of C) resides. The report is a central act of understanding between C and D; the report's value resides precisely in its ability to be silent about irritations and to talk about what has been known for long or is at least expected-without bringing in special learning stimulation that seems ‘out of context.’ Instead, a loose coupling between talk and action, in the sense of Nils Brunsson (2003), can be observed. Occasional understanding between these two organisations might even take the character of information hindrance and would then effectively become a learning barrier.
ARCHIVES
What the organisations learn in exchange with one another represents their growing capacity to operate in the DC sector. That which has been learned must be stored as sector knowledge. This function is fulfilled by organisational mechanisms which can be called archives.13 It does not matter whether such knowledge bases are stored and organised electronically, in paper form, or in another manner. Every organisation which exists in this kind of interorganisational relationship is in one way or another a ‘learning organisation.’ For the DC sector it is particularly important whether this learning is the result of an organised procurement of new information, or the inevitable, minimum-effort maintenance of established knowledge bases that hardly change. Learning with open objectives can become the precondition for a change in the status quo. But learning can also be adjusted to convenient methods and arrangements, in order to slow down or evade changes.
Figure 4.4 The importance of archives.
For organisations of type C two archives are particularly significant. A consulting firm collects all the reports connected with a project. This refers not just to the quarterly or to similar reports (inception report, rapport final), but also to those documents that had to be produced in connection with the acquisition of the project.14 These knowledge bases can occasionally simply be copied and used once again for new project documents, particularly in connection with acquisitions. This is a common practice as the conceptual targets in tender documents produced by D (as part of the learning relationship D >>> C not treated separately here) are based on the key terms15 already mentioned. And these key terms are to be found in the quarterly reports mentioned, which can now be recycled. Archives are not searched—and they would be in this sense hardly fruitful—for whether development aid projects have had no impact and have to be considered as failures. An open discussion on the basis of the existing knowledge bases is not carried out. Archives are museums for ‘trophies,’ with which a consulting firm communicates its functioning in comparable enterprises with the corresponding expert and country-specific conditions.
Just as important is the archive comprised of the curricula vitae of organisation members, of those permanently employed at a consulting firm or of those experts hired under temporary contracts. Such personnel constitute an essential component of a bid for a new development aid project during a call for tenders. The personnel archive is therefore much more actively maintained (and is entrusted to specially appointed members of the organisation) than the archive of project reports. One could say that the acquisition value of an expert grows almost exponentially in relation to the duration of his or her missions abroad. ‘Experience’ is thus rewarded here, too, and consists of learning ways of thinking, keeping deadlines, and using resources efficiently. There is no question that these are valuable characteristics and competencies for experts in a development aid project. No space remains, however, in these curricula vitae for indications of the ability or readiness to explore new paths, that is, for learning in its traditional meaning (cf. Weick 1996). DC is not an area where D, the provider of resources, wants to be unpleasantly surprised. This circumstance is of course known to C and R.
In this organisational context, archives function as strengtheners of the status quo, at least as concerns the area of consulting firms, and as a reassurance for those routines that characterise DC, contributing to an elaborate acknowledgement of “that's the way we have always done it.” It is not easy to take or suggest new learning paths away from the burdensome weight of these knowledge bases. At least this would not be possible without the willingness to take risks, and that is hardly to be expected.
THE ON-SITE MEETING
Since up to now we have only considered learning situations between two of the actors who are involved with learning in an organisation, the question might arise of whether a ‘shared learning process’ occurs in the form of the joint presence of actors from all the types of organisations at the same time and at the same place, or whether such a shared learning process can be triggered by a kind of ‘spontaneous’ encounter between the representatives of the different organisations involved. Such encounters are conspicuously rare. Two-sided learning situations clearly dominate, and could result in three-sided learning only through their concatenation across time.
The journey by representatives from all the types of organisations to the ‘place of the occurrence’16 represents such a three-sided learning event (linked with ‘practice’). Not in every case, but often it takes the following illustration: The D delegation would like to visit one of the 250 villages that profits from a hand pump project to supply drinking water. The village they will visit should be a ‘typical’ one. A convoy of five all-terrain vehicles sets off and after a one-hour journey arrives in the village. The village might be typical, but of course not the situation. The appearance of the village's inhabitants is by no means authentic, since the project administrators have announced the visit beforehand and an impressive performance has been arranged. Especially popular seems to be attendance of such a delegation to an education course where a ‘volunteer,’ trained in health and hygienic by the project, holds such a course for the entire population of the village, employing homemade didactic materials. A course with such good attendance has never taken place in the village before; and it will never happen again. The donor delegation, however, is satisfied and R and C are pleased that no ‘disasters’ have occurred. The convoy can then travel to the nearby provincial capital for lunch.
Figure 4.5 Three-sided learning event.
Nevertheless, even such a contrived event could offer enough material for learning-relevant irritation. The delegation could, for example, discuss with C and R the basis for the volunteer's actions17 during and after the project. The representative from R could explain the specific political reasons for accepting this specific village in the hand pump programme, but not the much needier ones in the surrounding area.18 The project head could raise the issue that the modest share of the investment costs, which should have been provided by the village inhabitants, could not be collected, in spite of all the efforts made, and that according to the project's rules, the project was actually not allowed to equip the village with a pump.
Three-sided encounters do not have to take place on-site, but can occur just as well in the conference room of a ministry. They are most often characterised by a broad disregarding of issues that could potentially build up a certain argumentative pressure to change the ways of decisions-making and acting. Although, as mentioned, three-sided encounters are rare, they are nevertheless—particularly and also precisely because of their scarcity—paradigmatic for learning obstruction, and for learning prevention processes in the field of DC. The fear and flight from irritation dominate the scene here in a particularly striking manner.
THE MANAGEMENT OF TERMS
Another typical learning situation concerns the processing of information in the recipient organisations. The learning recipient organisation is affected by the flow of information19 from D and C. Projects are not isolated schemes. They have developed out of other schemes, and although they are meant to solve problems or address target groups, they should also have beneficial effects for the donor organisation. Against this background, R takes over nolens volens legitimating patterns of argumentation from its ‘partners’: from D as refers mainly to the strategic targets, from C20 to the practical management of the process. The representative of R who masters the necessary key terms and the line of thought connected with them, and knows how to enrich them with stories about incidents s/he has experienced, demonstrates that s/he is an expert. Key terms include for example ‘participation,’ ‘decentralisation,’ or ‘good governance.’ R can signal to D and C that understanding will be successfully achieved.
Such a flow of information, whose novelty value quickly diminishes, could in principle flow in the opposite direction, from R to D and C. In view of the distribution of resources and of the idea that it is R, or the field of action represented by R (not D or C), which should develop, this, however, remains a purely theoretical possibility. Judging this constellation might seem difficult, but in the sense of ‘self-determined learning’ it is not the case. And it increases the probability that the more powerful such a superstructure becomes, the more the objectives and methods focus on sustaining this very superstructure.
A mode of speaking utilising project-sensitive key terms and the corresponding ‘thought’ can be observed not just at the higher echelons of a recipient organisation responsible at the national level. The same phenomenon occurs just as much in non-governmental organisations, which can be conspicuously close to governments, and among those members of the target groups who have learned to fulfill more or less profitable functions (such as ‘agents’) in various projects.
Figure 4.6 Behavior of the recipient organization.
A climate of prefabricated discourse facilitates—particularly from the viewpoint of C—the efficient realisation of a development aid project, but hampers the profound understanding of specific conditions and specific project effects. Above all, in such a mental framework failure (and thus the form of learning based on it) is almost impossible; the focus lies on the non-discussable key terms. In the best of cases only tactical learning can occur, a kind of mimicry whose range is limited to the respective occurrence of that project. While D certainly does not want to be irritated by any inappropriate facts in the report written by C (learning relationship C >>> D), in the learning relationships to D and C sketched out here R functions as the ideally irritated one, who feels obliged to state: “Well, that's the way we'll do it now!”
THE DC PROJECT AS PARADOXICAL TRANSNATIONAL ENSEMBLE
On the whole the phenomenon of ‘development aid project’ proves to be a highly differentiated hybrid construction. Development aid projects are themselves not organisations, but rather project relationships that extend broadly outwards, for which the term of transnational ensemble seems appropriate. As transnational ensembles, development aid projects are embedded in multiple and variable administrative or organisational-political constraints. These involve measures by organisations which hardly know, and are not able to assess, each other. Before they can reach objectives and target groups, they have to come into contact with each other. The cooperation between these organisations is successful precisely when their respective forms of existence and their particular logic are not fundamentally impaired through the contact, that is, when learning does not take place.
The framework for action of the phenomena designated as ‘projects’ is not produced by the projects themselves. The framework, moreover, cannot be derived from the events themselves; it exists instead as a multiple drawing of borders in the organisations involved. Development aid projects are not self-contained entities that speak for themselves, but rather the intersection of concerns that bring forth an organisational project occurrence in the narrower sense, whose basis is precisely not located in an independent project organisation. In contrast, a project distinct, ‘on-site’-focused, problem-solving concern on the part of the participating organisations and their individual actors is hardly evident. The occurrences constantly lack an overarching inner logic, with the weak inner drive corresponding to the numerous organisational boundaries and graduations.21
At the same time, development aid projects are transnationality in a paradoxical form. When one leaves the macro level and disregards the embedding of development cooperation work in a political ‘one world’ rhetoric, then it becomes clear that beneath the transnational surface it is organisations—and therefore intra-organisational viewpoints—that dominate. The transnational aspect appears from the perspective of organisations as quite an empty space, only corresponding to the inner logic of participating organisations to a limited degree. One encounters instead an ensemble of organisations, mutually creating the context for each other, which have to cooperate, but can only do this with tactical reservations. Reciprocal adjustment, not orientation, dominates here as the common relationship.
Under such conditions, the repeated call for ‘more transparency’ can find adherents in the participating organisations22 only to the point where this recommendation can be sensibly expressed according to the organisation's interests, or to a point from which transparency has to be literally hindered. When it already makes no sense for the organisations involved to grant reciprocal insights into each other, this applies all the more to the ‘empty’ transnational area mentioned earlier, whose illumination can only be disadvantageous for all the organisations. The limiting point is marked by the barrier of the functional lack of transparency: A complex ensemble such as a development aid project—as presumably DC itself—can only work when one lets the organisations involved have their organisational ‘private spheres,’ which in no case are capable of communalisation.23
In the DC landscape, therefore, a highly professional protection of the status quo flourishes, which hinders stringent learning processes and does not want to recognise the corresponding irritations. In its place, the smooth execution of projects and programmes becomes the uppermost precept of decisions and action. It is thus not surprising that the flows and concatenations processed in this ensemble for the most part evade simple (textbook) ideas about ‘learning organisations’ and ‘knowledge and learning based development aid.’24
On the whole, a rather pessimistic view of learning presents itself in the interorganisational functioning of DC, in contrast to the first optimistic model. The comparison of the two illustrations (Fig. 4.2 and Fig. 4.3) makes clear a cardinal difference: While in the optimistic variant progress in learning and knowledge is assumed for the sector in general (transferred to the ordinates), the more pessimistic and in our opinion more realistic variant displays a lateral movement (indicated by the arrows). There is no ‘general’ growth in abilities and knowledge.
This is noteworthy insofar as the knowledge bases (e.g., the archives) of the individual organisations (with the exception presumed here of only slight growth in R) clearly increase from project to project. However, these knowledge bases—aside from their possible irritating contents of project occurrences—are tailored to the respective functional necessities of the participating organisations (thus, in contrast to Figure 4.2, indicated with a striped surface).
Knowledge bases in individual organisations might increase, but the system as a whole does not experience a growth in knowledge. Transcending the borders of the organisation and taking in unfamiliar, complex process rules without endangering one's own organisation or other forms of more closely linked knowledge and learning processes have to be regarded as improbable. When both illustrations are compared, it becomes noticeable that the DC system under observation in the more pessimistic variant is substantially more simply structured. What is missing is the feedback of information to D and C from learning situations experienced by R. The system's discursive zone limits itself to D and C.
Figure 4.7 A more pessimistic view of interorganisational learning in DC.
WHAT HELPS THE SECTOR LEARN BETTER?
Interorganisational learning is—presumably not just in DC—an exceedingly contingent endeavour, which can easily become entangled in contradictions or remain trapped in dead-ends which at first seemed attractive. One can repeatedly observe that organisations do not use knowledge, which is actually available, nor do they adopt knowledge which is actually learnable, although they could act and make decisions in a more informed and efficient manner. This suggests by inference that for the organisations such knowledge can only be considered in a prepared or ‘cleaned up’ manner. The usual concept of a learning organisation would in fact be that of an organisation which continually irritates itself, which even undertakes that which in the long term it can hardly expect of itself without creating new structures. Such a concept might still work as a programme for small organisations, which operate in a context without strong institutions and with charismatic leadership. In a sector such as development cooperation, however, controlled by large organisations and institutions with a pronounced division of labour, opportunistic learning in the sense of adaptation to institutionalised patterns is much more probable. ‘Sectors that learn’—and the more complex the more so—are likely only to remain an idea. Evolutionary change might occur, but not that type of change which eliminates the fundamental problem. Interorganisational learning in transnational relationships or sectors which organise themselves is instead subject to the principle of trial and error, or remains limited to communities of interest which spontaneously arise, are strongly dependent on trust and successful reciprocity, and thus not very stable. In any case, it is a slower, much more fragile process than that in a well-oiled bureaucracy with optimised learning strategies. Against this background we would like to finally examine intelligent innovations and irritations that are capable of increasing the structural conditions for learning in such a transnational ensemble as well.
A sort of lodestar for intended change—and not just in the sector analysed by us—is the call for higher expenditure, in other words for larger budget shares for DC. Few people in the sector are likely to close to their minds to this demand. It seems, however, not very probable that an increased flow of funds will enable the organisations involved to change their manner of learning. On the contrary, it is more likely that, ceteris paribus, the ‘margin for error’ will become larger, thanks to reduced intra-sectoral shortages and competitive situations.
Similarly, the improvement of evaluations25 or the increased professionalism of the evaluating personnel would initially have one effect: the boom in the sector function of ‘evaluation.’ It is unlikely that such evaluations would penetrate the organisational structures of the sectors in the sense that these structures—at least partially—are recognised and regarded as superfluous, and instead the needs of the target groups are more clearly considered as the central premise. It would be even more improbable to expect from the efforts put into evaluations anything other than conclusions, and not wide-reaching consequences. Evaluation and learning remain—whether with ‘quality improvement’ or not—a self-referential act among the organisations operating in the DC sector, since these spare each other mutual impositions as far as possible.26 The currently highly favoured strategies of context control, in the sense of good governance, likewise do not appear very promising in light of the actual existing organisational structures. We suspect that interorganisational learning will free itself even more from replicable project strategies and more detailed process knowledge. Instead, learning itself can focus more on the maintenance of the context as well as on the demonstration of the ‘context ability’ of the participating organisations.
As regards an answer to the question posed in connection with Thiel's (1998) observations mentioned earlier, whether the premises identified by him could point the way to a development aid practice that learns better, as of now a few cautious assumptions can be put forward. Our diagnosis in this regard is quite sceptical. A sector divided into many players cannot raise well-functioning projects in the sense of a general applicability to a binding ‘curriculum,’ nor can one expect an eye-opening function from ‘clear writing’—what would the actors then see that they cannot already see? It would be more decisive—however (un)clear—to write about the facts beyond the organisational horizons of expectation.
Similarly, we are quite sceptical as regards the strategy of strengthening the learning ability of the DC sector by organisational re-arrangement of the development agencies, as it is difficult to understand how organisational arrangements of this kind could avoid the logic and the attachment to the sector. Moreover, as the ‘aid octangle’ and the illustration of knowledge accumulation (Figures 4.1 and 4.7) suggest, it is simply not enough to locate independent learning, innovative concepts, strategic courage, etc. in the donor organisations. This should be done instead just as much in the other organisations involved.
In our viewpoint a more promising premise seems to lie in the reorientation of the donor organisations. Development banks influence the thinking of the entire sector and force learning processes to be oriented towards their perspectives and process rules in the spirit of administrative uniformity. This not only makes DC prone to errors, but also administratively complicated. Development banks should perhaps limit themselves more to that which they undoubtedly do best—namely provide resources (conveniently according to the stipulation of long-term, fixed country quotas)—and content themselves with a function as payment and auditing office.
In this case, the generation of ideas and concepts would have to take place somewhere else, in the developing countries themselves. As long as this remains undone in a country or in a sector, this would be understood as a sign that (till now) no substantial demand for the corresponding aid exists, at least none for which a multiplier effect of the development aid project would have a high probability. The demand for resources which has to be linked with such an arrangement should obviously not be expressed haphazardly. The demand for resources articulated on-site should, quite differently from up to now, be directed at an expert competence also located on-site, broadly as independent from the recipient as from the donor organisations. These could be organisations with the will towards organisational independence and a certain code—perhaps such as the Red Cross. Learning and knowledge could then be possibly organised in a manner closer to problems and target groups. On the whole, diversity and ideas competition would noticeably increase. In addition, the apparent failure of concepts necessary for learning progress would have a stronger chance of being possible and recognised.
NOTE
1. This chapter presents a substantially reworked and enlarged version of Ehlers and Wolff (2008).
2. SADEV: Swedish Agency for Development Evaluation
3. “e+z” is published together with a parallel edition in English: “development and cooperation (d+c)”
4. Bundesministerium für Wirtschaftliche Zusammenarbeit und Entwicklung [German Federal Ministry for Economic Cooperation and Development]
5. By assuming a core of eight types (the category “Civil Society Organizations” appears once on the donor as well as the recipient side) of aid organisations and involving the actual beneficiaries only as the ninth actor, Ostrom et al. (2002) already imply that the flows of information proceeding from and to the actual beneficiaries (represented as dotted lines) are not indispensable for the functioning of the total system (which may have to do at least as much with the way ‘functioning’ is defined as with whether the system ‘functions’ or not).
6. This division of functions can assume different forms: Non-governmental organisations (NGOs) can take over the functions of administrations and consulting firms, often simultaneously; donor organisations have their own in-house consulting departments; and organisations like the Catholic Church may sometimes assume all three functions in one project.
7. We are concerned here with one selection including the three types of organisations described; we do not claim to reproduce the totality of typical learning situations in DC.
8. In the four development aid projects described in this chapter, reports from the project head level (consulting firm) to the (superordinate) level of the donor organisation constitute a fundamental part of the processing of the project. The reports are expected to accurately and sufficiently reproduce the project occurrences within the period under review. There are regular progress reports as well as a variety of special reports (start reports, interim reports, and final reports). The rule is no report, no project, since the donor organisations can proceed with the necessary actions (for example, payments, authorisations, or approvals) only on the basis of such reports.
9. This report is usually passed on to the recipient organisation (R) or agreed upon with it.
10. The term ‘irritation’ is used here as an analytical tool. Irritation is not used in order to value or appraise something. It indicates a for the moment undefined perturbation which is not compatible with a personal or social system's established expectations (Luhmann 1997: 789 ff.). Effective irritations induce a higher attentiveness but may also lead to surprise, annoyance, or even vexation. Depending on the system and its current condition, an irritation may be very welcomed as ‘good news’ or firmly refused as “out of standard” (Luhmann 1995: 63). Karl Weick uses in this respect the terms ‘interruption’ and ‘recovery’ (thereby referring to John Dewey (1922: 178-79)): “Order, interruption, recovery. That is sensemaking in the nutshell” (Weick 2009: 39). Hence, organisations learn less from ‘irritations/interruptions’ than due to them (cf. Christianson et al. 2009), provided the ‘window of opportunity’ is wide enough open to new knowledge and patterns, i.e., to learning (Tyre and Orlikowski 1994). All in all, the capacity to deal with irritations is a necessary condition for learning.
11. The Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ, German Society for Technical Cooperation) employs such terms. ‘Progress report’ or ‘rapport d'avancement’ are common. GTZ has recently changed its name to Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ).
12. A fascinating episode along these lines is described by Rottenburg (2009).
13. The accompanying sub-illustration cannot be directly derived from the main illustration (Figure 4.2). The concentric circles depict the growth in archived knowledge in the learning relationship C >>> C from the time points t1 to t2.
14. Generally, the ‘technical’ (expert) bid and the financial bid.
15. C-personnel pay close attention when D-personnel employ key terms in new or repeated ways or noticeably often.
16. As a potential arena for all six conceivable learning relationships, namely D >>> R, D >>> C, R >>> C, R >>> D, C >>> D, and C >>> R.
17. Volunteers who represent a certain promise of sustainability customarily also profit in material terms from projects (in the form of mopeds, T-shirts, or payments) when they make substantial contributions to the project's progress.
18. Doing this would be risky for the R representative; moreover, the questionable practice would continue.
19. Learning relationship D >>> R and C >>>R
20. The organisations encounter each other in a project occurrence via the functionaries typical for projects. On the side of the consulting firm this is the project head, at the donor organisation this might be the country expert, and in the recipient organisation there is the figure of the counterpart.
21. Naturally this does not imply that development aid projects have to always be insipid, or that projects do not lead to results—only that these results often appear to a certain extent ‘unofficially’ (that is, not mentioned, or only incompletely mentioned in the reports).
22. James March (1994: 64) has no illusions about this as regards other organisations: “The belief that more information characterizes better decisions and defensible decision processes engenders a conspicuous consumption of information. Information is flaunted but not used, collected but not considered.”
23. To the extent that sector communication can lead to decisions—for example, in the form of resolutions—then what March (1994: 195 ff.) has described would be valid: “Participants may share some objectives, but characteristically their coalition is a negotiated coalition of convenience as much as it is one of principle. . . . Disagreements are resolved by vague language and vague expectations. . . . Decision makers interested in building viable coalitions are likely to seek and find allies who will be vigorous in supporting symbolic decisions and lax in implementing them.”
24. At least it is likely improbable that the requirements of reflection enumerated by Bierschenk (1997: 12) are often achieved: “Thus, whether the actors involved in it are conscious of it or not, development assistance—be it on a national, macro- or on a local, micro-level—is always and first and foremost development policy, and even development politics . . . If every development initiative is an intervention in a complex social game then it might well be worth to find out some essential facts before embarking on it: What is at stake? What are the rules of the game? Who are the main players? Which side do we want to be on? Who will profit from our presence? What is the likely price, in sociological terms, of interfering in the on-going game? Who are we going to hurt?”
25. On evaluations in DC see Schaumburg-Müller (2005).
26. Luhmann (2000: 330 ff.) fittingly writes of the “poetry of reforms.”
BIBLIOGRAPHY
Bierschenk, T. (1997) ‘APAD 1996: On the move. . . . ‘ In Le développement négocié: courtiers, savoirs, technologies, Bulletin N° 12, edited by T. Bierschenk and P. Le Meur, 1–5. Hamburg: LIT Verlag.
Brunsson, N. (2003) The Organization of Hypocrisy: Talk, Decisions and Action in Organizations. Copenhagen: Copenhagen Business School Press.
Cassen, R., and Associates (1986) Does Aid Work? Report to an Intergovernmental Task Force. Oxford: Clarendon Press.
Christianson, M. K., M. T. Farkas, K. M. Sutcliffe, and K. E. Weick (2009) ‘Learning Through Rare Events: Significant Interruptions at the Baltimore & Ohio Railroad Museum.’ Organization Science, 20: 846–60.
Dewey, J. (1922) Human Nature and Conduct. Mineola, NY: Dover.
Ehlers, K. E. (2011) Projekte der Entwicklungszusammenarbeit. Kooperation und Abgrenzung in einem organisationalen Ensemble. Hamburg: Kovac.
Ehlers, K. E., and S. Wolff (2008) ‘Grenzen interorganisatorischen Lernens. Beobachtungen aus der Entwicklungszusammenarbeit.’ Zeitschrift für Pädagogik, 54 (2008): 691–706.
Ingram, P. (2002) ‘Interorganizational Learning.’ In The Blackwell Companion to Organizations, edited by J. A. C. Baum, 642–63. Oxford: Blackwell Business.
Krohwinkel-Karlsson, A. (2007) Knowledge and Learning in Aid Organizations. A Literature Review with Suggestions for Further Studies. Karlstad: SADEV.
Luhmann, N. (1995) ‘Die Behandlung von Irritationen: Abweichung oder Neuheit?’ In Gesellschaftsstruktur und Semantik. Studien zur Wissenssoziologie der modernen Gesellschaft. Vol. 4, 55–100. Frankfurt: Suhrkamp.
Luhmann, N. (1997) Die Gesellschaft der Gesellschaft. Frankfurt a. M: Suhrkamp.
Luhmann, N. (2000) Organisation und Entscheidung. Opladen, Wiesbaden: Westdeutscher Verlag.
March, J. G. (1994) A Primer on Decision Making. New York: Free Press.
Ostrom, E., C. Gibson, S. Shivakumar, and K. Andersson (2002) Aid, Incentives and Sustainability: An Institutional Analysis of Development Cooperation. Sida Studies in Evaluation 02/01. Stockholm: Swedish International Development Cooperation Agency (Sida).
Ramalingam, B. (2005) ‘Implementing Knowledge Strategies: Lessons from International Development Agencies.’ ODI Working Paper 244. London: Overseas Development Institute (ODI).
Rottenburg, R. (2009) Far-Fetched Facts: A Parable of Development Aid. Cambridge, MA: MIT Press.
Schaumburg-Müller, H. (2005) ‘Use of Aid Evaluation from an Organizational Perspective.’ Evaluation, 11: 207–22.
Scott, W. R., and J. W. Meyer (1991) ‘The Organization of Societal Sectors: Propositions and Early Evidence.’ In The New Institutionalism in Organizational Analysis, edited by W. W. Powell and P. J. DiMaggio, 108–40. Chicago/London: The University of Chicago Press.
Thiel, R. E. (1998) ‘Der geheime Lernprozess.’ E+Z—Entwicklung und Zusammenarbeit, 12: 307.
Tyre, M. J., and W. J. Orlikowski (1994) ‘Windows of Opportunity: Temporal Patterns of Technological Adaptation in Organizations.’ Organization Science, 5: 98–118.
Weick, K. E. (1996) ‘The Nontraditional Quality of Organizational Learning.’ In Organizational Learning, edited by M. D. Cohen and L. S. Sproull, 163–74. Thousand Oaks: Sage.
Weick, K. E. (2009) Making Sense of the Organization. Vol. 2, The Impermanent Organization. Chicester: John Wiley and Sons.
Wright, J., and M. Winter (2010) ‘The Politics of Effective Aid.’ Annual Review of Political Science, 13: 61–80.