Journal

• Editorial Board
• Subscription
• Table of Contents
• Urban Morphology online
• MS Submissions
• Join ISUF

Journal - Viewpoints vol.8 nr.1 (2004)

• K. Kropf - M.R.G. Conzen, Gianfranco Caniggia, Oscar Wilde and Aesop
• M. Gallarati - Urban-scale architecture today
• R. Williamson - What's left out?
• K.-J. Kim and P.J. Larkham - Reply to Williamson

M.R.G. Conzen, Gianfranco Caniggia, Oscar Wilde and Aesop: or, why urban morphology may be right but not popular

Karl Kropf,
Roger Evans Associates, Kidlington, UK and Urban Morphology Research Group, School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham B15 2TT, UK. E-mail: KarlKropf@rogerevans.com

Having been introduced to the work of M.R.G. Conzen and Gianfranco Caniggia as a student I subsequently spent time following up the sense that they had much in common. Since then I have tried, in turn, to introduce their ideas to a variety of people in a variety of ways. There have been times in attempting to explain those ideas and discuss them that I could not help thinking of a quotation from Oscar Wilde: `When people agree with me I always feel I must be wrong' (Ellman, 1968).

The hare of fashion and the tortoise of understanding

Not, of course, that Wilde doubted his own view. Beyond the diffident waspishness, one can sense a firm belief that the popular and easily grasped - the fashionable - must in some sense be wrong, or at least not the full truth. Which is, I think, one of the more fundamental things that Caniggia and Conzen had in common: a firm belief in the truth - or at the least, persuasiveness - of their view in the face of fashions. On reflection, Wilde's droll quip and its relation to `more serious' matters such as ideas about buildings and towns shows a marvellous intertwining of views. Is he serious or is he not? I like to think of it in terms of Aesop's tortoise and hare. Is Wilde a hare, only interested in passing fashions, or a tortoise wanting to get to the bottom of things - or an inextricable combination of the two? Stretching just a little further, is the inextricability part of the truth? Fashion, as a phenomenon, is not a superficial issue. It is not something that will just evaporate. It may be infinitely flexible and resilient but it is not insubstantial and can present a barrier to ideas. Both Conzen and Caniggia in their careers worked in a context in which fashions in ideas presented a significant obstacle with which to contend.

For Caniggia, it was, amongst other things, the idea of the star-architect, architecture as a channel for personal expression and pure innovation and the seemingly infinite expansion of the conceptual determinants of form. These sat in clear opposition to the Muratorian idea of the architect as a technician of the built environment.

While the realms of architecture and urbanism would appear to be more prone to fashion, Conzen, as a geographer, was no less subjected to it. Just as Conzen published his seminal work on Alnwick, the world of human geography was leaving historico-geographical urban morphology in what many considered to be a backwater.

Not that typology and morphology have lacked their moments of popularity. The 1980s were the glory years with the Krier brothers, Aldo Rossi and Anthony Vidler, amongst many others, sunning in the light of attention that had turned from an exhausted Modernism. But it is perhaps because of that brief moment of fashionability, along with the longer-term effort to take the ideas to a non-professional audience, that Oscar Wilde's quote came to mind.

Putting it in more explicit terms, I began to wonder why I thought that urban morphology and building typology may be right but not popular. In digging a little it seemed to me that there are a number of core ideas that are like `eternal mysteries', ideas that, on the surface, appear simple and straightforward but, on closer inspection, seem to evaporate. Perhaps not surprisingly, it is those core ideas that are shared by Caniggia and Conzen.

The eternal mysteries of urban morphology and building typology

Their work is fundamentally based on the notion of class or type. In consequence it involves probabilistic and population thinking and, by extension, statistical inference. In particular, to remain unambiguous, it is important to avoid making category errors between the class as a whole and an individual that is part of the whole.

Conzen's and Caniggia's work is based on relations between things rather than `real objects'. A common strategy is to define entities by relative position. People, in general, prefer to think in terms of real, solid, `objects'. In particular, both Conzen and Caniggia identify a set of relations that form a nested hierarchy. There are `parts', `parts of parts' and `parts of parts of parts'.... Dealing with this in the abstract can, at the most basic level, lead to difficulties of language.

Caniggia's and Conzen's approach is based on the notion of process or relations of continuity through time. If spatial relations are intangible, relations through time are more so given that they are not `present'. On top of that, the notion of process identified by Conzen and Caniggia incorporates the auto-corrective or self-regulating circuit - circular chains of determination - the extension of this is the homeostatic system and `mutually modifying habits'. In such systems it is difficult (makes no sense) to identify any unambiguous linear sequences of cause and effect.

The processes that Conzen and Caniggia identify fundamentally involve chance and probabilities. The patterns that constitute the subject of urban morphology and building typology are emergent. They cannot be deduced from initial conditions.

The resulting package of ideas leads to a `subject' that is intangible, complex, historical, paradoxical and equivocal. It is not obviously aesthetic. It does not appeal to the appetites but does, perhaps, to ideas of truth and veracity.

To risk a crude generalization, things that are popular in a wide sense are: tangible, simple, present, unambiguous and permanent. They appeal to the appetites and may contradict ideas of truth and veracity. The appeal of these latter attributes in many ways lies behind the impulse to build: to create something with such qualities - tangibility, simplicity, presence, unequivocality and permanence.

The inescapable chestnut of description and prescription

But at this point it is important to make a distinction. There are two realms in which urban morphology and building typology are applied. On the one hand there is the realm of description and explanation: understanding how things work, their `handling characteristics' or `behaviour'. On the other hand there is the realm of what people want or desire - what they seek to achieve when they build. Those two realms are separated by a philosophical chasm. This is the old `description/ prescription' chestnut.

The apparent paradox of saying urban morphology and building typology may be right but will not be popular is, in part, resolved by recognizing these two realms. Typology and morphology may be right in their description and explanation but will not necessarily be popular as a basis for prescription.

The only way to cross from one to the other is human choice, exercised individually or corporately and always within a specific, contingent and conditioned context. There may be obvious or compelling choices but there are no necessary choices.

Conzen and Caniggia did, of course, make a choice. Both saw a significant value in the notion of continuity and of ensuring the historical process that led to the physical structure and character of the built environment remained evident in it. They sought to learn from the historical process and use that learning in creating new environments or renewing old ones.

In this vein Conzen put forward the idea of townscape management. In his view, under-standing the historical process of development and its manifestation on the ground in plan units and geographical regions was essential background information. With that information it would be possible to manage change in a measured way to retain continuity and at the same time allow for emerging needs.

Caniggia elaborated the design method of reprojection by phases, essentially modelling the local, historical process of change to arrive at new designs. He took a design through phases of transformation up to a point and proposed buildings that he expected would be the starting point for further transformations.

But even here there is another paradox - the reflexive paradox. The move from description to prescription involves a change in context, a change in level from looking at the process as a whole to performing a single step at a particular point within the process described. You end up looking at yourself acting in the process. Yet it is not always an obvious jump and one that is often disguised by the rhetoric used in taking a particular position regarding design. Choices are presented as necessities, just as often to justify change (we must evolve or die...) as to justify continuity. Again, Caniggia and Conzen both advocated continuity but were aware of the jump in level. Of the two, Caniggia was particularly aware of the difference. His design method is predicated on consciously designing as if he were not conscious of the process as a whole but acting in accordance with a locally generated tradition of building.

Slow and steady

All of this suggests that, while fashions may be an obstacle from a particular point in a longer term process, they are less important to the process as a whole. Does it matter in the long run that urban morphology and building typology are not popular with most architects or house

builders? The conclusion of Aesop's fable of the tortoise and the hare would seem to provide the answer.

In the long run it would seem more important to ensure we get the descriptions and explanations right. What has been accomplished so far is immense and I never cease to be amazed by the work done by Conzen, Caniggia and those who preceded and followed them. The job is, of course, never done. We need to elaborate and refine our understanding of the built environment by challenging received ideas, presenting new ones and continuing to test them all.

We also need to extend the range of ways we have to promote designs that make use of that improved understanding. To do so involves ensuring our focus is not just on the so-called traditional and historical but also on the contemporary, keeping up with and seeking to understand the current context in which designs necessarily fit and participating in current lines of typological processes. This, in turn, means engaging with the `industry' as it stands, with, in most places, housebuilders and developers, with planning officers, local government represen-tatives and the general public. My own experience in this respect is positive. People do not have trouble `getting' the core ideas of morphology. It is too easy to underestimate people's capacity. People do, however, get put off by jargon and tend to turn off when things get too complex.

But more complex ideas have become commonly understood. Rather, one might say, misunderstood. That will always be the danger. Complex and equivocal ideas that fit the facts (because the facts are complex and equivocal) will always be open for interpretation. From another point of view, it should be a positive sign that people dispute the ideas. If, like Oscar Wilde, we are wary of someone who too readily agrees with us, we might welcome an informed and constructive dispute.

Reference

Ellman, R. (ed.) (1968) The artist as critic: critical writings of Oscar Wilde (University of Chicago Press, Chicago).

Urban-scale architecture today: a new town centre near Genova

Mario Gallarati,
Dipartimento di Progettazione Architettonica, Università di Genova, Stradone S. Agostino 37, 16123 Genova, Italy. E-mail: m.gallarati@iol.it

The study of Italian architectural piazze published in Architettura a scala urbana1 offers a fresh view of the more or less well known examples of the forum-square from the Renaissance and the later, larger-scale baroque interventions. At the same time, the study contrasts the responses, in different situations, to the question of urban space treated as an architectural whole.

In the case of urban spaces created both within existing tissues as well as within new settlements or new quarters, the theme of the forum-square remains current. As an intervention, it requires confronting problems varying from the urban scale (the role of the public space within the urban organism) to the scale of the building (definition of the public space by a number of individual buildings) and the architectural scale (with particular reference to the treatment and articulation of the building façades).

The study and interpretation of examples from the past offers the possibility of not merely documenting a particular aspect of the history and morphology of a place, but also identifying type-solutions for recurrent problems of architectural composition. As such, the study becomes a critical-operative instrument that is indispensable for the planning and implementation of new complexes of buildings and architecturally unified urban spaces.

An experiment in this sense has been undertaken by the author through the design and realization of the new town centre of Casarza Ligure, near Genova (1991-2003). In this case, the study of the past has not resulted in proposing formal-stylistic elements borrowed from another epoch, but in acquiring the necessary awareness of the different problems connected with a project, the relevant type-solutions that constantly present themselves in architectural projects at the urban scale, and their reinterpretation in the light of present-day exigencies and building techniques.

The relation to the urban organism

The first aspect to consider concerns the position of the public space (street, square) within the system of axes and poles defining the settlement. There are two main cases. In the first, the square is built within an existing tissue, through restructuring, usually with the aim of enlarging and regularizing a pre-existing nodal space (see, for example, the Piazza Ducale in Vigevano, Piazza del Popolo in Ascoli Piceno, Piazza of S.S. Annunziata in Florence or the Piazza del Popolo in Fermo). In the second case, the urban space is a central unifying element among new buildings: historically, at first, as a court of honour facing castles or palazzi (see Piazza Grande in Carpi, Piazza Bentivoglio in Gualtieri and some of the Gonzaga squares), then as axes and poles of new expansions (as in the case of baroque Turin).

The new town centre of Casarza Ligure has been constructed on an open area beyond the river that faces the old centre, to which it is connected by a bridge. In the plan, we have tried to take into account those factors in composing, around an architecturally defined central space, an urban tissue which, starting from the incidental character of the existing periphery, comes to acquire an appropriate degree of organization in becoming the centre of a new settlement (Figure 1).

The spatial system and its relation to urban tissue

Taking as a starting point the historical examples, which have been significantly constrained by the existing fabric, the interventions seeking to systematize public spaces always tend towards solutions that are more organic and regular. In particular, the uniformity of squares derived from the restructuring of existing urban tissue in many cases consists of the mere superimposition of homogeneous building façades on the hetero-geneous tissues that define the space (again, examples include Vigevano, Ascoli Piceno, Carpi and Fermo).

In the case of new building, however, the public space organized along an axis or system of axes, is defined by serial building fronts that are typologically homogeneous and architecturally unified (see again S.S. Annunziata in Florence, in

particular the building of the Loggiato and Serviti houses; also Gualtieri, the Gonzaga squares of Pomponesco, Rivarolo Mantovano, S. Martino dell'Argine and, created later, the seventeenth- and eighteenth-century squares in Paris and Turin).

Other public spaces, including some of the most significant, occupy an intermediate situation, their ultimate arrangement derived from successive phases of construction, tending towards symmetry and polarized by the presence of the church (as in Vigevano and Florence).

The new quarter planned for Casarza Ligure presents a regular urban tissue with blocks of buildings analogous to those of the surrounding area, developed along two main axes: a longitudinal axis polarized by a central square delimited by two building façades with arcades, for residential and commercial use, and a new cross-axis on which stand the buildings facing the river (Figure 2). The façades of the serial buildings of the square, which are lower than the the residential blocks at the back, have a role of mediation between the main public space and the new residential building tissue.

The architecture of urban space

As we pass from the typological-building scale to that of the formal-architectural, the variety of solutions adopted increases, according to the different contingent situations, the period of origin and the local building vernacular. But even in this case, there are some recurrent problems of composition and the relative type-solutions adopted that can be singled out.

First, the rhythmic front is the formal result of modular components organized serially as part of the building façades defining the urban space. The module, whether it appears as an elementary three-dimensional organism (as in Vigevano) or is limited to the mere thickness of the front, generally coincides with the span of the bays on the ground floor and a corresponding arrangement

of windows in the upper storeys. The archi-tectural unity of the façades is pursued through the repetition, in an individualized rhythm, of that elementary module. The rhythm may, on the one hand, be uniform, with bays of the same width (the most common solution), in some cases accentuated in the upper storeys (either by doubling the openings as in Procuratie Vecchie in Venice or by alternating chimneys and pinnacles as in Gualtieri). On the other hand, the rhythm may be articulated by alternating modules of different width (as in Piazza Castello and Piazza S. Carlo in Turin).

Secondly, the entrances to urban space or the theme of access to the square (closely connected with the theme of termination and the return of the rhythmic wall into the streets leading into the square) is itself dealt with in very different ways. The more elementary solution is the mere interruption of the arcade, which therefore presents its side to the roads leading into the main urban space (as in Ascoli Piceno, towards the secondary roads, in Carpi, or in Vigevano, towards Corso Vittorio Emanuele II and Via XX Settembre). A more elaborate solution is turning the arcade along the main routes of access. This was done in Ascoli towards the Trivio, where there are only three bays facing the Loggia dei Mercanti; in a more conscious and systematic way, in the Gonzaga squares of Pomponesco and S. Martino dell'Argine; and, above all, in Turin, in the great seventeenth-century developments subsequent to the original composition of Piazza Castello. A further step is the building of a true gateway, either shaped as a triumphal arch, as theorized by L.B. Alberti (which we still find in the sixteenth-century square of Isola Dovarese and was perhaps present in the original settlement of the Vigevano square), or as a clock tower (for example, the Codussi tower of Piazza S. Marco in Venice and that of Gualtieri, standing in the middle of the western prospect of Piazza Bentivoglio).

Thirdly, the presence of a special building, which is itself the cause and justification of the architectural square, assumes a variety of forms. These range from the incidental location of the special building in the square (as in Ascoli, mainly with respect to the church of S. Francesco, which pre-dates the square and turns its side towards Piazza del Popolo) to the more or less co-ordinated insertion of the building within an arcade (as in the case of the churches of Pomponesco and Gualtieri), the nodal positioning of the special building at the end of the main axis of the spatial system (for example, the church of S.S. Annunziata in Florence, Palazzo Bentivoglio in Gualtieri or the Gonzaga castles, later demolished, for which the square was a court of honour) to, finally, the placement of special buildings along the main accesses to the square (like the twin churches of Piazza S. Carlo in Turin).

Lastly, there is the matter of architectural apparatus which, by its nature, is the theme of the project. The architectural unity of façades made up of a number of contiguous buildings leads to a simplification of the decorative apparatus and to an accentuation of the horizontal elements meant to make the comprehension of the whole more immediate. Regarding the former, responses range from an enquiry into simple geometric regularity (as in Vigevano) to the repetition of standardized details (for example, the white travertine windows of the square of Ascoli Piceno, and the mouldings and the serena stone windows and the majolica tondi in the Florentine square), the framing of the bays of the portico inside an order (as in the Piazza del Santuario di Loreto or in the Portico dei Banchi in Bologna and, later, in Fermo or Isola Dovarese) and the simplification and superimposition of the order reduced to mere framing (as in Pomponesco or in Gualtieri). Regarding the latter, the responses vary from the simple continuity of the gutter line (as in Vigevano) to the introduction of stone window cornices (as in Ascoli), and the prog-

ressive introduction of storey-marking cornices functioning as trabeation (as in Florence, Fermo, Isola Dovarese) and of unified gutter cornices.

In the case of the new urban centre of Casarza Ligure, the whole project is based on a modularity of alternating structures and voids. This choice is reflected not only in the typology of the residential buildings but also in the alternating rhythm of the walls defining the central square. Corresponding to the different width of the bays of the portico there is a variation in the treatment of the fronts of the upper storey, with tripartite loggias corresponding to the minor cells and a pair of windows related to the major ones. The arcade turns with two bays extending along the main axis out of the square, towards the old centre. It finishes at the opposite end with two special buildings either side of the same axis, characterized by greater height and co-ordinated with the rest of the architecture of the square. The treatment of the main façades, plastered in the traditional Ligurian manner, and the continuity of the gutter cornices help to define the architectural unity of the entire building complex.

Note

1. Gallarati, M. (1994) Architettura a scala urbana: urban scale architecture Studi e Documenti di Architettura (Alinea, Firenze).

What's left out?

Rebecca Williamson,
University of Illinois at Urbana-Champaign, Study Abroad Program in Versailles, 2 Avenue de Paris, 78000 Versailles, France. E-mail: rebecca.williamson@wanadoo.fr

The ISUF Conference in Trani was my first exposure to the organization's activities. A year earlier I had participated in the Landscapes of Water Conference (also organized by Attilio Petruccioli and the Politecnico di Bari) in the town of Monopoli in southern Puglia, and had been expecting something similar: a heterogeneous mix of people from widely scattered countries and disciplines, thrown together in a seemingly random fashion, their initial lack of commonality ultimately won over by the warmth of Puglian hospitality and the urgency of the problems at hand.

The Trani conference was similar to the Monopoli conference in many respects, but there was an important difference. Unlike at Monopoli, at Trani it was clear from the start that a specific set of methodologies was assumed as the framework within which everyone either worked or did not. That framework relied primarily on traditional plans and charts comparing measurable aspects of the city such as density of inhabitation. Both adherents and non-adherents seemed frust-rated: the former for having to listen to the `irrelevant' papers of those adopting other approaches, the latter because of the barriers to communication that this split caused.

Although spontaneous connections began to occur as the conference progressed (again, the effects of Puglian hospitality and urgent problems), I believe this issue warrants consideration precisely because the challenges our cities face are so serious. Groups such as ISUF, who wish to have an impact on the urban experience, must find the most effective ways to communicate - to transmit, yes, but also to receive - across disciplines, across nationalities, and across methodologies. To cite just one example of a potential area for methodological exchange, among the presentations at Trani was one describing the way in which sound articulates space in Jenne, a city in Mali. The traditional tools of morphological analysis fall short in conveying phenomena of this kind, and must be complemented by other forms of research.

In my own experience of the Trani conference one of the most memorable moments was the presentation of the International Urban Form Study. There the contradiction so often associated with the urban morphology `movement' - the tension between description and prescription - was apparent, yet not fully addressed. To a large extent this resulted from a well-meaning desire for an apolitical approach.

The ambitious study, eloquently described by Paul Hess in the last issue of Urban Morphology, compares the urban forms of Seoul, Tokyo, Paris, London, and Los Angeles. The announced goal of the study is to inform policy-makers in Seoul about how their city compares with other cities of similar size, particularly with regard to density. Every attempt has clearly been made to ensure a rigorous, unbiased comparison among the selected cities. The methodology is consistent across all the cities: each researcher sampled diverse areas of the city and avoided extreme or atypical conditions. The research was presented primarily through measured plans and charts depicting density considered according to a variety of criteria.

The presentations provide an opportunity to question whether one can ever truly view the city in a neutral way, or whether, in attempting to evacuate subjective, `non-scientific' considerations from the analysis, one creates a void into which rush unintended political, social, and other influences.

While the study adheres to the standards of academic research and is thus necessarily abstract and constrained, it is at the same time directed toward a specific practical situation with its attendant complications. It is in this dual nature that the study holds both promise and risks. In seeking to apply academic research to an existing situation, the project crosses a threshold from disinterested comparisons to potential benefits and losses for people living in the city now and in the future.

I should note here that I have never been to Seoul and have thus far experienced Korean culture primarily gastronomically, through acquiring an appreciation for that cuisine's craving-inducing blend of pungent flavours. My comments are, therefore, those of an outsider both to Seoul and to ISUF, and are based in more general questions about the methods we use to study and shape the city.

The following are some of my concerns:

"The researchers have chosen only cities that they consider to be `good' examples; places that Seoul would presumably like to resemble in some way. Cities they considered to be `bad' examples, such as Mexico City, are not part of the study. It would be useful to know the criteria used to judge whether a city is a good or bad example, and to have some negative examples to weigh along with the positive.

"The list is weighted toward former imperial powers, including Japan, Korea's occupier during the first part of the twentieth century. This raises questions about Seoul's own self-image and aspirations.

"The study examines only cities about which legible information was readily available. Cities either not documented in this way or with unwilling governments were not considered. This limited the study to cities with a specific kind of political structure. The possession of quantitative information on the part of a government regarding its citizenry and territories is not, however, simply a neutral, apolitical fact. Quantifying acts such as census taking and cadastral mapping had to be invented and maintained, as they were in Europe over the course of the past few centuries. If, for example, one compares a Napoleonic plan to a Korean plan of the same period, one can see that France and Korea saw themselves in different ways even at a relatively recent phase in the development of the two cultures.

"The research promises an objective and scientific approach to the city, and resembles a taxonomic study. Certain qualities escape the rigorous framework of the study, including the effect of climate, customs, etc. on the experience of a specific configuration. Similar spatial configurations in different places might be experienced in disparate ways.

"The study succeeds in debunking some myths about the relative densities of contemporary cities, but what precisely does this give Seoul? Whose interests are served as a result?

The presentations of the research in Trani reminded me of a French saying: `Paris est Gruyère'. The saying reflects the way that both the city and a French Gruyère cheese reveal cavities when seen in section. Could not all cities be seen, perhaps, as cheeses, or even as tofu, some ripe and smelly, some fresh and bland, each best appreciated when cut with a particular knife, in a particular shape or direction, taking into account the relative qualities of core and edge? One can `sample' a cheese and compare it with another in terms of its quantitative aspects: its moisture and fat content, for example. This information is useful for regulators and nutritionists, but will never really tell what is distinctive about that cheese. I would argue that the same holds true for cities.

The authors of this impressive study could further purge the study of subjectivity in search of purer neutrality; or they could acknowledge the political dimension of the exercise. I hope that they will choose the latter. In the process, they could expose ways in which political, social, and other qualitative and experiential aspects interact with measurable aspects of urban form to make the experience of a specific city what it is. Seoul would have no easy answers as a result, but might find in such a study the basis for a probing consideration of what that city is and wants to become.

Notes

1.Holder, G. and Olivier, E. (2003) `Jenne: music as a political act in an African Muslim city', in Petruccioli, A., Stella, M. and Strappa, G. (eds) The Planned City? (Uniongraphica Corcelli Editrice, Bari) vol 2, 682-8.

2. Kropf, K. (2003) `M.R.G. Conzen, Gianfranco Caniggia: fashions and mysteries', in D'Amato Guerrieri, C. and Strappa, G. (eds) Gianfranco Caniggia: Dalla lettura di Como all'interpret-azione tipologica della città (Mario Adda, Bari) 59-64.

3. Scott, J.C. (1998) Seeing like a state: how certain schemes to improve the human condition have failed Yale Agrarian Studies (Yale University Press, New Haven).

Reply to Williamson

Kwang-Joong Kim,
Seoul Development Institute, 391 Seocho-dong, Seocho-ku, Seoul 137-071, Korea. E-mail: kjkim@sdi.re.kr and Peter J. Larkham, School of Planning and Housing, University of Central England, Perry Barr, Birmingham, B42 2SU, UK. E-mail: peter.larkham@uce.ac.uk

The contents of ISUF's meetings, projects and publications, including this journal (not least the publication in it of Williamson's viewpoint), speak for themselves about the willingness to receive, transmit and engage with a wide range of views, methods, disciplines and nationalities. However, on behalf of the research team, the specific points that Williamson makes about the Seoul study need to be addressed.

"It was not the case that only cities considered `good' were selected. Some of the environ-ments examined possessed characteristics that many would regard as undesirable. In residential areas, for example, the compari-sons demonstrated some high densities, inadequate vehicular and pedestrian circu-lation space, and little private or public open space. The use of `good' and `bad' by Williamson seems to imply value judgements about the quality of the cities that were certainly not made by the research team. Our primary concerns were pragmatic and methodological. What physical forms should be examined in a comparative study of urban form? What data are required, and how can suitable data sets be obtained?

"There were, therefore, pragmatic aspects to the selection of cities. These included the timescale and budget, which were set by the funding body. Issues such as contacts with suitable scholars, their availability, and the availability of appropriate data sources were equally important, but had nothing to do with the Seoul city government. Nor was the final city selection related to Seoul's self-image or aspirations. Seoul is, however, the primate city of the world's eleventh-largest economy. It seems appropriate, therefore, to make comparisons with other major cities that have experienced fast growth phases. That three of the six chosen cities are in former colonial countries is coincidental: as far as we know, neither Sedoul nor South Korea have colonizing aspirations!

"The research clearly demonstrated that international comparisons are (a) at a very early stage, (b) complex and difficult, and (c) demand considerable effort in setting-up the study and in face-to-face meetings and workshops. To make this effort more manageable and comparable, cities with accessible data sets were selected. One important and salutary lesson from the project is that, even so, data access can be exceptionally difficult and/or incomplete (for example in US cities immediately after 11 September 2001).

"Taxonomy is often indicative of an early stage in a discipline's development. ISUF is a young organization, even if some of urban morphology's roots can be traced to the last century but one. There is still much to learn from each other, and thus taxonomic and comparative approaches seem to have much utility. Inevitably, however, the pragmatic

elements of research - time and budget - precluded any systematic consideration of other aspects or qualities such as Williamson mentions. That is not to say that the research team feel that these are irrelevant.

"It is true that Seoul wanted to make comparisons between some selected aspects of urban form demonstrated by cities of so-called `advanced' countries. In many respects, the development of Seoul's urban form follows their experiences. The study has also been shaped by the interests of its co-ordinating body, the Seoul Development Institute. The emerging concept of urban sustainability suggests the need for a reassessment of different urban forms and densities (and, ultimately, experiences and other qualities) of different cultures, and thus a comparison of large cities in Asia, Europe and North America has been an increasingly important theme in seminars, research and publication in urban planning, geography and related fields. It is as a result of this interest that the Seoul Development Institute initiated this study as a relevant and practicable early step.

All research is, in some sense, political and pragmatic. We had hoped that these limitations to the Seoul project had been sufficiently explored in the presentation and subsequent discussion at the Trani conference.

It now seems appropriate to move on to explore other points about the study - and possibly about comparative study more generally. What factors impede such studies, and how might they be overcome? How can we develop research proposals, and assemble research teams, that will attract substantive research funding? How can such morphological research inform practice?