Beyond the Paradox of Hierarchy*

Dr. Tatsuro Sakano, Associate Professor

Graduate School of Decision Science and Technology

Tokyo Institute of Technology


Centralization contradicts decentralization. So does hierarchy vs. network.However, they are practically wrong when seen from the history of organizational development. If I call these phenomena as the paradox of hierarchy, how to understand the nature of this paradox is the key to understand spatial impacts of Information Technology.

Drifting Among Reality, Ideal, and Illusion

In five yearfs time, all companies will be Internet companies or they wonft companies at all. This prophecy is quoted in Economist of June 26th 1999 as the words of Andy Grove, the chairman of Intel.It is also said that only a few companies can survive in one business sector on Internet Market.As always in the booming periods, the psychological pressure have forced us to rush into Internet companies and Dot Com businesses before the market is amazoned.[i]  

As mass market was created with the development of railway system since the late of 19th century, newly developed telecommunication technology has been creating a huge market at global scale.However, even in the U.S., one of the most leading countries of e-commerce, the total spent by online consumers still amounted 1% of all retail sales (Economist, Feb. 26th, 2000).Most of fast growing net companies still remain deficit.It is very difficult to tell reality from ideal and ideal from illusion especially in the midst of rapid change.

M. Castells (1996) discussed that the world took off from industrial to informational mode of development around 1970s.A Japanese economist Yujiro Hayasi wrote a book titled gInformational Societyh in 1969, when D. Bell (1973) characterized the newly emerging society as post-industrial society.It is still debatable whether this societal change in the last three decades of the 20th century is to be called informational or post-industrial.[ii]The nature of informational revolution and its impact on spatial configuration are still vague and ambiguous.It is required to see the phenomena in a longer and historical perspective for avoiding confusion among reality, ideal, and illusion.

Technological Impacts on Spatial Order through Organizing Principle

The relationship between technology and societal change is interdependent. Technology does not deterministically lead to societal transformation.[iii]  Society will not change or the existing regime will be strengthened by new technology if it does not fit to the social system requirement.Spatial impact of information technology depends on the structure of industrial organization.To understand the organizing principle of industrial organization is the key to understand technological impacts on spatial order.

Although industrial revolution is usually looked upon as a consequence of technological revolution, it only tells a half of the story.  Potential power of new technology can not be fully attained without organizational revolution.The organizational revolution, which occurred at that time, was the bureaucratization of society.[iv]Right now, in the beginning of the 21st century, we are still in the process of searching a new organizing principle in the emerging informational age.Although the answer is not yet clear, there is a general consensus on the point that the age of bureaucracy is over. 

From a marginalist economic perspective, organization, whether public or private bureaucracy, is considered to be inferior to market.[v]  Public choice economists criticize the innate nature of self-interest behaviors of bureaucrats, which distorts the appropriate supply of public goods.[vi]The organization studies, which include management science and organizational sociology, have been exploring the alternative principles to bureaucracy.[vii]  Finally, in stead of the coercive, centralized, decision making, participation has been claimed for in various political movements.[viii]As G. Peters (1996) pointed out, although their reasons to fight against bureaucracy are different and there is no unified theory, all of the recent reforms are based on a common assumption, the deficiency of bureaucracy.[ix]Especially, centralized control structure through formal rule and hierarchy are considered to be the obstacles to innovativeness and flexibility.[x]In its stead, network and decentralization are considered to be the promising new organizing principle.

The implications of this organizational transformation and the prediction of its impact on the spatial order were dissolution of the concentrated metropolitan cities and recreation of small communities.P. Goldmark (1972) predicted that gtelecommuting would liberate people from the conditions of extremely density within the confine of cities.h[xi]  We can find the same idea in A. Tofflerfs (1981) gthe electric cottageh, which is conceived glocationally liberated from the need to concentrate in cities.h[xii]These early writers predicted spatial decentralization as well as social and economic.

However, more recent writers, for example S. Sassen, M. Moss, and M. Castells argue for the concentration of creative activities by the empirical studies from the late 70s through 80s.Sassen (1991) revealed the concentration process of world big corporationsf command functions into a few gglobal cities,h which occurred since the late 70s.Moss (1986) also found that information intensive industry, producer services in particular, had been concentrated in a few global cities further than before, and especially in the C.B.D.s within these metropolitan regions.Castells (1989) argues persuasively that this kind of concentration is not limited to the producer service but the same phenomena can be seen in innovative manufacturing activities.Castells calls this special place ginnovative milieu.hPlace has still important roles to nurture creative activities as against the expectation of the early, futuristic, predictions that all the information transaction can disappear into electronic network and dissolve the importance of nodal places.In effect, the world city system constitutes hierarchical order based on the command and control hierarchy for the entire global economy.

The idea that the city system is hierarchical can be traced back to W. Christallerfs (1933,1966) gcentral place theory.hThe new version of the hierarchical theory of the informational age can be found in the work of J. Friedmann (1986) and D.A. Smith & M. Timberlake (1995).If hierarchy is the reality, are network and decentralization illusions?Or are network and decentralization the leitmotif of the informational society in a longer perspective and hierarchy and centralization in the last three decades just a swing back phenomena in a short term?

Myth of Silicon Valley and Network Economy: An Alternative to Hierarchy?

The U.S. of 1990s has marked a clear contrast to the 1970s and 80s sluggish economy.The symbol of the new economy is Silicon Valley.  A. Saxenian (1994) analyzed three major factors of the success; local institution and culture, industrial structure, and corporate organization. The image of Silicon Valley, she depicted, is a region with risk taking culture, accumulation of high tech small firms, and flexible network of them.  From J. Gottmann (1967), J. Jacobbs (1965, 1984), M. Moss (ibid.), and M. Castells (ibid.), these prominent urban researchers recognize that active person to person communication or improvisational human interaction are the sources of inventing new ideas.It is not impossible to coordinate creative interactions beyond geographical boundary through computer network, if an organization has enough human resources and if they are hierarchically organized.In the long run, however, this hierarchical network will loose to the open interactive places like Silicon Valley because hierarchically controlled knowledge creation network is organized for keeping their inventions secret from the outside of the organization.

On contrarily to the Saxenianfs image, B. Harrison (1994) pointed out that most of the small firms do not afford high technology, low paid, and are positioned in a peripheral part of the hierarchically organized global production network, from the center of which a few world corporation command and control.So which version of Silicon Valley is the reality not the illusion?The debate concerning hierarchy vs. network still continues.

The Paradox of Hierarchy

Since 1950s, two principles of modernization, positivistic rationality as a legitimate mode of knowledge creation and bureaucracy as an efficient organizing principle, have been seriously questioned.Since then, the most intellectualsf preferences have tilted towards postmodern principles, that is, network, decentralization, and diversity, as against modernistic principles, that is, hierarchy, centralization, integrity and universal rules.

But is hierarchy really obsolete and ought to be taken over by new organizing principle?Reviewing the history of organizational development, we notice that the reality is more complex than the world the theories presuppose.First of all, standardization has been spreading further and further at the same time diversification proceeds.[xiii]It is not difficult to find the examples of this phenomenon. For example, diversified software run on a few standardized OS.We depend on manuals while we pursue diversified customer satisfaction by one to one marketing. The unique value of ethnicity and local culture is emphasized at the emergence of a unified global culture.

Secondly, we think strong leadership is necessary for quick and fast management at the same time delegation of power is demanded from the same reason.[xiv]  The merger of big companies is booming at global scale and keeping positions in the center of global network while outsourcing auxiliary functions and creating small companies network. [xv]  The top 500 TNCs (Trans-national corporations) generate 30% of gross global product at the starts of 1990s (Myers, 1994, excerpted in Graham and Simon, 1996).Among top 1000 world financial institutions, the top 25 share the 30% of the entire asset in 1990. This share has increased to 39% in 2000 (Asahi News Paper, August 19, 2001).Finally, as Graham and Marvin (1996) pointed out, mobility in time-space depends on spatially fixed and embedded accumulation of physical constructions and networks.The flexibility in the cyber space is actually realized on the wires and optic fibers, which is spatially fixed and embedded.

Standardization contradicts diversity by their definitions. So does strong leadership vs. participation, centralization vs. decentralization, hierarchy vs. network, and mobility vs. fixity.These definitional contradictions may seem to be logically true.However, they are practically wrong when seen from the history of organizational development.Lawrence & Lorsh (1976) discussed that high performance organization develops high integration mechanisms to keep balance with high differentiation.In theoretical biology, Bertalanffy (1950) noticed a transition toward states of higher order and differentiation in organic development and evolution.I would like to call these phenomena as the paradox of hierarchy, a paradox of societal integrity vs. individual freedom or that of centrifugal vs. centripetal forces. 

How to understand the nature of this paradox is the key to understand new design principle of organization and newly emerging spatial order. 

Beyond The Paradox of Hierarchy

The paradox hierarchy implies that we should not see the system, whether it is an organization or spatial configuration of cities, with the dichotomy of hierarchy vs. network or that of centralization vs. decentralization.The development of organization and that of the spatial order can be achieved by the shift from lower level equilibrium of centrifugal vs. centripetal forces to the higher level equilibrium.Hierarchical structures will never disappear.So the right question is not to ask whether network takes over hierarchy or not, but to ask in what way a transition toward states of higher order and differentiation is proceeded. Historically speaking, construction of urban places has been devised to overcome time with space (Graham and Marvin, ibid.).In pre-modern times, physical closeness makes communication easier.  In modern era, the development of transportation expands this function of cities.In the same way, telecommunication technology has now shifting cities in the higher order of time-space compression.The cities, which established global link either by Silk Road, by ocean route, by rail track, or by air route, used to be wealthy cities.Now, the cities are linked by telecommunication technology, thus achieving the higher level equilibrium of centrifugal vs. centripetal forces. 

In computational science, it is suggested that almost all the complex system adapt hierarchy as a fundamental design principle to cope with combinatorial complexity.[xvi]Hierarchy is a useful strategy to overcome the limit of our computational capability.  Modernistic design principle on the other hand is based on the perfect rationality.In M. Weberfs terminology, calculability is the essence of formalistic rationality.Rational knowledge means calculability.Control by rational knowledge, embodied by experts as a formal rule, is the core of bureaucracy.The application of the modernistic principle in the field of city planning is the spatial regulation by standards and rules.

Questioning about bureaucracy as a legitimate organizing principle, which starts around 1950s, is rooted to the doubt about perfect rationality.Strategic planning is proposed to overcome the deficiency of pre-set rules and manuals. Epistemologically speaking, it is M. Polanyi (1958) who proposes gpersonal knowledgeh, which can not be known as explicit propositions but known tacitly.Just like a piano player can play Bach, but he/she can not explain explicitly how to play.The player knows how to play piano tacitly but not explicitly.Nonaka (1990) analyzed the limit of bureaucracy and the strength of Japanese management by focusing on tacit knowledge.

Tacit knowledge is the foundation of most of our activities.It can only be transferred through face to face communication.In D.B.Audretschfs terms (1997), the marginal cost to transfer tacit knowledge still increases with the distance while it becomes independent with the distance because of telecommunication technology.This is why innovative activities need face to face contacts, while routine work can be controlled by explicit knowledge beyond the distance, and why the role of place is still important.The recognition of the importance of tacit knowledge may have some connection with the S. Freudfs study on unconsciousness.Metaphorically speaking, bureaucracy controls human activities by societal consciousness.On the other hand, network is expected to utilize societal unconsciousness.However, we still do not understand well how to utilize societal unconsciousness.This is the challenge for us to invent a new way toward states of higher order and differentiation. 

*This paper is prepared for 37th IsoCarp Congress in Utrecht (16-20 September 2001).


[i]The verb gamazonh is derived from a business is done through Internet just like, the corresponding business sector is called amazoned.

[ii]The analytical framework proposed by Castells (1996) is helpful to understand the nature of the confusion between informational and post-industrial.He argues that the society should be characterized by two axes the mode of production (capitalism, statism) and the mode of development (industrialism, informationalism).In his definition, todayfs world is informational capitalism.However, capitalism is sometimes considered almost same as industrialism.This is the major source of confusion.
[iii]For this discussion, see Castells (op. cit.) and also Ackoff (1974).
[iv]Reorganizing social systems under the guide of end-means rationality has been called modernization.Bureaucracy used to be considered as a vehicle to achieve this rationality (Clegg, 1990).This tenet of the industrial age is expressed theoretically by Max Weber. Weber classifies the rationality into two types: formal rationality and substantive rationality.The dominant mode in the modern world is formal rationality.It is T. A. Parsons (1947, p.16) who redefined the formal rationality in terms of end-means relationship.In this paper, the term end-means rationality is used for the clarity of the meaning.See also the discussion by S. R. Clegg (1990).
[v]  See for example Hayek (1944) and Clegg (op. cit.)
[vi]  See Niskanen (1971) and Peters (1996)
[vii]  Following Woodward (1965) and Burns & Stalker (1961), contingency theorists have done a series of empirical studies (Donaldoson, 1995).
[viii]  The study on workers participation started by Michigan school as job enlargement from 1950s.These studies become the basis for the development of participatory management (Davis and Taylor, 1979, introduction).
[ix]G. Peters (1996) identifies four deficiencies of bureaucracy and corresponding visions of governance.These four emerging visions of governance are named as market government, participative government, flexible government, and deregulated government.They are designed to overcome deficiencies of monopoly, hierarchy, permanence, and formal rule respectively.
[x]  The characteristics of bureaucracy are understood slightly different among researchers.In Aston group of organizational study, which conducted empirical studies on organization since 1950s, mentioned five characteristics: specialization, standardization, formal rule, centralization, and hierarchy (Pugh, 1988).Weberian bureaucracy can be considered as a refinement of Saint-Simonean centralized hierarchical organization by introducing standardized formal rules.Taylorfs Scientific Management is an actualization of Weberian bureaucracy or it can be said that the latter is a theorization of the former.
[xi]This statement is quoted by S. Graham and S. Marvin (1996)
[xii]This statement is also quoted by S. Graham and S. Marvin (op. cit.)
[xiii]Flexible manufacturing system is based on standardization of parts.See the strategy of nature mentioned by Koestler (1967)
[xiv]A good example of this confusion can be seen in the recent government reform in Japan.It aims at strengthening leadership of the Prime Minister, while decentralization law has been passed.
[xv]Harrison (1994) calls this phenomenon as concentration without centralization. Castells (op. cit.) calls it decentralized concentration. 
[xvi]It is H. A. Simon (1969) who explained why most complex systems are organized hierarchically.It is J. Holland (1975) who explained the same phenomena from a computational perspective.An overview of their theories and their implications for organizational design are discussed in Sakano (2000).


Ackoff, R.L., 1974, Redesigning Future, John Wiley & Sons 

Asahi News Paper, August 19, 2001, gConsolidation of Euro-American Financial Institutions.h

Audretsch, D., B., 1997, gAgglomeration and The Location of Innovative Activity,h 

Oxford Review of Economic Policy 14

Bell, D., 1973, The coming of post-industrial society, Basic Books

Burns, T. B. & G. M., Stalker, 1961, The Management of Innovation, Oxford University Press

Castells, M., 1989, The Informational City, Blackwell Publishers Ltd

Castells, M., 1996, The Information Age: Economy, Society, and Culture, Vol. I, II, III, Blackwell

Publishers Ltd 

Chandler, A. D., 1962, Strategy and Structure, The MIT Press

Christaller, W., 1933,1966, Central Places in Southern Germany, Prentice Hall

Clegg, S. R., 1990, Modern Organization, Sage Publications

Davis, L. and J. C., Taylor, (ed.), 1979, Design of Jobs, Goodyear Publishing Company

Donaldoson, L., (ed.), 1995, Contingency Theory, Dartmouth Publishing Company

Economist, June 26th, 1999, hA Survey of Business and the Internet,h p.3

Economist, Feb. 26th, 2000, gE-Commerce: Shopping Around the Web,h p.5

Friedmann, J. 1986, gThe world city hypothesis,h Development and Change, 4

Gottmann, J., 1967, Megalopolis: the Urbanized Northeastern seaboard of the United States

Graham, S. and S. Marvin, 1996, Telecommunications and the city, Routledge

Harrison, B., 1994, Lean and Mean, Harper Collins 

Hayashi, Y., 1969, gJouhouka Shakai (Informational Society),h Kodansha

Hayek, F. A., 1944, The Road to Serfdom, Routledge and Keagan Paul

Holland, J. H., 1975, Adaptation in Natural and Artificial Systems, The University of Michigan

Jacobbs, J., 1961, Economy of Cities

Jacobbs, J., 1984, Cities and the Wealth of Nation

Koestler, A., 1967, Ghost in the Machine, The Macmillan Company

Lawrence, P. R. and J. W. Lorsh, 1976, Organization and Environment: Managing

Differentiation and Integration, Harvard University

Moss, M., 1986, gTelecommunications and the future of cities,h Land Development Studies 3

Niskanen, W., 1971, Bureaucracy and Representative Government, Aldine/Atherton

Nonaka, I., 1990, Knowledge Creating Company (Chishiki Souzou no Keiei),


Parsons, T., 1947, Max Weber The Theory of Social and Economic Organization,

The Free Press

Peters, G., 1996, The Future of Governing: Four Emerging Models, University Press of Kansas

Polanyi, M., 1958, Personal Knowledge, The University of Chicago Press

Sakano, T., 2000, eDesign Theory of Planning Organization: An Computational Approach,f in 

Y. Kumata (ed.), Kokyo Sisutemu no Keikakugaku, Gihodoshupan (in Japanese)

Sassen, S., 1991, The Global City, Princeton University Press

Simon, H. A., 1969, The Sciences of The Artificial, the M.I.T. Press

Smith, D.A. and M. Timberlake, 1995, gConceptualizing and Mapping the Structure of the World

Systemfs City System,h Urban Studies, Vol.32, No.2

Toffler, A., 1981, The Third Wave, Morrow

Von Bertalanffy, L., 1950, eThe Theory of Open Systems in Physics and Biologyf, 

Science, vol.111, pp.23-9

Woodward, J., 1965, Industrial Organization, Oxford University Press