The central importance of road hierarchy is that it has the ability to influence the relationship between different types of roads and the overall structure of the urban layout. So it's a fairly abstract concept, but it can have very concrete consequences, as we've analyzed in the previous chapters, in terms of the road hierarchy, in terms of the destruction of the urban integrity by cutting off the urban highway, and in terms of the auto-oriented urban landscape. The traditional road hierarchy seems to be incompatible with contemporary new urbanism. Because it seems to prioritize cars, few explicitly encourage public transit. And the new traditionalists favor connecting transportation networks. Negate any important role of traditional city streets. It's a negative connotation, and it's associated with modernism and engineering.
As we have found, architects and planners under the influence of the modernist model use roads and their hierarchical structures for structural purposes, and isolate nearby buildings according to their functional and usage street. In addition, contemporary designers of new urban design may propose, just as they advocate the clear hierarchy of space, a certain hierarchy, which will play a positive role in the formation of a certain mechanism. Therefore, it is closely related to urban structure to analyze the problems of road structure hierarchy and operation mode. The traditional road hierarchy, while retaining the word road hierarchy, cannot be denied the possibility that streets or other roads are accommodated. How to formally classify the rank system and define the relationship between different road types and the relationship system with the rank road is the focus of our discussion.
5.1 Hierarchical 'Trees'
The modern road system usually focuses on the hierarchical division of roads, in which the discrete route types are clearly identified and the relationship between them is controlled. The road's functional hierarchy stems from a need, road or traffic network that combines vehicles with pedestrian and non-motor vehicle route systems to provide an effective motion function in a separate design. Influenced by this view, any motor vehicle traveling between a specific starting point and an end point has as little intrusion as possible into adjacent areas and living areas.
The traffic hierarchy, mainly to avoid conflict, not only to avoid conflict between efficient traffic flow functions, but also to ensure safety and comfort among traffic participants, and to improve the environmental quality of urban areas. So designers do consider non-traffic factors in urban environments, although this may still be quite limited in terms of the broader possibilities of urban design.
5.1.1 Of Traffic, Towns and Trees
The hierarchy of traffic structure has great influence on the design and management of road network.
In the analysis of the basic principle of Buchanan urban structural unit method, it is found that the urban corridor plays the role of network connection to the environmental area. So we see the release of a series of problems, such as isolation, which is a restricted connection channel, at the lowest level of the hierarchy, as in the form of a tree. This creates two types of roads. The perfect system assumes that the tree road will constitute IsdK7t/GcyHVivoXW0re9w==a relatively high proportion of the road condition. This led to the most secondary roads being considered fit as the front passage. Usually any kind of urban road, including those that form the main street, is divided according to its functionality. That is, the influence of roads on people, buildings and other urban functions is gradually reduced.
However, this analysis method plays a negative and hindering role in the sustainable development of cities.
Two kinds of roads
In the play beauty scenario, the tree-structured road would constitute a relatively high proportion of the road with only secondary roads considered suitable for positive access.
In the system envisaged here, distributor roads would form a relatively high proportion of all roads, and only the most minor roads are deemed fit for frontage access. Typically, any sort of through road in an urban area - including those forming local high streets – would tend to be categorized according to its function as a road - ie, as a distributor, thereby relegating its importance with regard to people, buildings and other urban functions.
This has often resulted in urban blight, as routes designated as primary distributors are progressively 'improved' by straightening, requiring demolition of property, and obstruction or discouragement of frontage uses. In fact, it is clear from the examples in Traffic in Towns - and consideration of any real case - that the retrospective categorization of existing roads in urban areas is fraught with difficulty and ambiguity, requiring the forcing of all sorts of road into arbitrary categories based on supposed traffic significance and reinforcing these distinctions by subsequent remedial action or improvement.
It seems that there are simply not enough categories of road, as presented above, to reflect the rich complexity of road and street types which exists on the ground. Even in the scenario above, where a single distinction between distributors and other roads is attempted, the need for subdivision of the 'other' category seems inevitable.
A further objection to this hierarchy is that while distributors may be designed for movement, there is no reference to the means of movement envisaged. Undifferentiated motor traffic appears to be assumed, which not only omits pedestrian movement, but also fails to differentiate treatment of vehicle types such as cyclists or service vehicles, or the sharing of road space with other modes such as trams.
Indeed, the road hierarchy promoted in Traffic in Towns is almost antithetical to the objectives of public transport orientation. This is because it effectively separates public transport from its users (pedestrians) and from those users' origins and destinations (urban 'streets', shopping areas, houses, etc) by separating routes for traffic distribution, on which the buses would travel, from those for pedestrian use and access to building frontages.
The mega-structure of Traffic in Towns' Fitzrovia case study is a particularly stark example, where the buses are separated from the pedestrian deck by up to two escalator flights , but the basic problem of spatial separation still applies in more mundane examples of 'prairie planning' up and down the country.
It may be concluded that, in order to satisfy contemporary objectives of sustainable mobility and urban vitality, any hierarchy must take account of transport mode and frontage development, and allow combinations and interfaces which promote the greener modes and the use of public space.
Nomenclature
Traffic in Towns suggests that only three types of distributor road and one type of access road are recognized:
We think this comparatively simple nomenclature could with advantage replace the present large number of terms - arterial roads, through roads, expressways, freeways, principal traffic roads, collector roads, service roads etc. - which are freely used with little if any standardization of meaning.
Aside from the Orwellian overtones of this deliberate contraction of the language, there are objections to the very logic of the argument: the problem alluded to is not the diversity of terms, but the lack of standardization of use, and yet the proposed solution is to reduce the diversity. It is argued here that a more diverse - albeit standardized - nomenclature could assist the promotion of a more diverse route typology which could better accommodate a variety of transport modes and urban functions.
The tree analogy
This innocuous-looking statement arguably goes to the root of the problem of 'disurban creation'. Although superficially a simple analogy, it manages to combine no less than five distinct connotations in one. These connotations are worth exploring in detail as they help illustrate some fundamental points which distinguish important aspects of urban structure:
This tree analogy encapsulates an important distinction. It demonstrates that different types of road may be distinguished, ranked in importance, and given allowable relationships between them. This amounts to typology and hierarchy, which together shall be referred to as the constitution of urban structure.
The analogy also deals with the relative number and organization of connections of each type. This gives rise to the network topology, or configuration of the structure.
However, there is a problem here in that, in the end, it is not clear just how many of the connotations are to be taken literally. Effectively, the problem of 'Buchanan's Tree' is due to the confusion of constitution and configuration. Buchanan's tree analogy may not only be problematic because it limits the allowable connections between different route types, but because of its association, however unconscious, of such a kind of hierarchy with tree-like layouts terminating in a multitude of 'twigs' - that is, cul-de-sac.
5.1.2 Hierarchy and layout
Hierarchy and layout are often confused: the terms are often used ambiguously and apparently interchangeably. Thus, the term 'hierarchical' is commonly used to denote a 'loop and cul-de-sac' layout, even though other potential patterns, including traditional grids, may also be hierarchical in nature, either by design or retrospective designation.
For example, separate traditional suburban development from traditional community development, linking the former with a "layered street network" and the latter with a "reduced street level." In the sCGd2woIgtxHiVzTGncbf/g==ame way,
Attributed the "hierarchical street network" to traditional suburbs rather than the "highly connected grid streets" of new traditional communities. Refers to "an integrated grid network in contrast to a highly differentiated street hierarchy." Here, a special grid is envisaged and urged to get rid of the "hierarchy". However, the layout and hierarchy seem chaotic here. Adopting a particular layout, such as a grid, does not preclude retaining a particular type of hierarchy. In particular, the kind of "integrated grid networks" envisioned by advocates of new traditional urbanism actually have a greater distinction between route types, which would help form more complex hierarchies.
At the same time, the graphic comparison between "dense networks" and "sparse hierarchies" is made. Since the latter can be interpreted as a "layered" layout, the word "sparse" here seems to actually refer to connection or routing density. Again, the hierarchy and layout seem to be confused.
It turns out that modern road layouts, combined with typical features such as circular dealer roads and dead-end lanes, not only design explicit hierarchical intentions, but these seem to be more hierarchical structures. This is because they seem to embody and replicate the concept of the hierarchy through a tree layout of ground road types.
As previously pointed out, the traditional road classification usually describes the road as a trident tree system, with each road taking traffic on less important roads and directing it to more important ones. Although this happens frequently, there is no explicit requirement to adopt this particular layout in hierarchical systems.
And so it can be seen that network type cannot be defined by hierarchy alone, as a range of different patterns, from grids to cul-de-sac arrangements, could be generated using the same hierarchical rules and relationships. Indeed, it is precisely because of this flexib0be137210bbd7a29b6258e5751aa9d2ble nature that it has been possible to apply modern hiee716efdfcd38d38e745d6523ac635512rarchical designation to whole road networks, with the retrospective application of hierarchical designations to ancient pre-motor age street forms.
Similarly, road type must be seen as being independent of hierarchy. Although for some hierarchical systems it is assumed that there are fixed relationships between road type, hierarchical level and transport mode (segregated motorway; local road; pedestrian access street) this need not be so. Indeed, it precludes the creation of many permutations of road types which although not explicitly catered for in conventional hierarchies - have a useful role to play in an urban road network, such as the 'mixed traffic urban Street' or multi-purpose boulevard.
The discussion so far suggests that some sort of modification or improvement to conventional hierarchy would be beneficial, in order to accommodate the ends of urban planners and designers and the reality of extant examples of traditional route types. We shall discuss the implications of this for the urban street shortly. In the meantime, though, we have another tree analogy to deal with.
5.1.3 A city is not a tree
In his influential essay A City is Not a Tree, Christopher Alexander explores the structure of cities in terms of sets of nested or overlapping elements, explaining and illustrating the types of structure known as the tree and the semi-lattice.
Alexander suggests that traditional towns and cities growing up in an organic manner result in semi-lattice structures which are rich in complexity and variety. He argues that these are preferable to more simplistic, planned settlements which are typically tree-like in structure and which are characterized by the separation of urban functions and which result in a loss of vitality. In Turner's words, Alexander emphasized that cities are not hierarchies, and that when planners believe they are, they produce the horrors of 'planned towns' with road hierarchies, business areas and useless open space.
Alexander argues that the temptation to design cities as trees should be resisted, and that instead they should be designed as semi-lattices. While the precise significance of the semi-lattice is open to question, the main point taken here is that a certain kind of hierarchical structuring, exemplified by the tree, has a resonance with the simplistic compartmentalization of functions often found in planned settlements.
This compartmentalization is perhaps seen most graphically in its spatial manifestation, in the plan view of new towns, divided into discrete neighborhoods . However, the functional compartmentalization also applies in the case of the separation of the roles of the traditional multi-purpose street (e.g., movement channel, public space, retail frontage) into separate, mono-functional elements (e.g., distributor road, civic piazza, shopping mall). The crucial point of this case is that the corresponding public uses of the traditional street， say, passage, promenade and purchase ，are mutually supportive. By contrast, a 'planned solution', in which these are separated, runs the risk of being dysfunctional.
On top of this, there is the connotation that, in some way, the fact that these traditional city elements and structures evolved together somehow influences their successful functioning. Thus it is not only the final product that is in question, but the process that generated it. This implies that even the 'multi-purpose traditional street', as a conscious design target, might be dysfunctional, if imposed in an appropriate manner.
We can use the term " Alexander's Tree " (referring to the tree that Alexander opposes) to refer to the top-down structure of urban areas that prescribes hierarchical relationships in an organized and simple way, which usually leads to the creation of disurban. Alexander's "tree" refers to the relationship prescription from top to bottom of tGzmV2FSsPgvPluiaYONbCuYz6/AXQZJGgzNmMrq95mQ=he planning process and the physical product represented by a single functional area or space.
Alexander's tree resonates with Buchanan's, though it is conceptually distinct. Alexander's Tree contains general methods of urban system planning, which can be easily applied to traffic planning and land use planning. Buchanan's tree is a particular way of linking network configuration (topology) to composition (road type and hierarchy). Both, however, deny the complex relationship like a tree.
5.1.4 A street is not a twig
It is suggested here that there is a parallel to be drawn between road systems – as significant generators of urban structure - and cities as a whole, in that planned road systems, in terms of road type and network pattern, typically have less complexity and diversity than traditional road systems. This might in turn contribute to the resulting settlements having the lack of 'coherence' and vitality of traditional urban areas.
It is part of this investigation, therefore, to explore how planned road systems，based on conventional concepts of road hierarchy - differ from the road systems of traditional urban areas, and in what way Alexander's Tree and Buchanan's Tree have contributed to disurban creation. One such way is seen clearly in the case of the traditional urban street.
In Traffic in Towns, the reference to the trunk, limbs, branches and twigs of a tree appears to refer to the progression from most important elements to the least important elements. However, as we have seen, it is also possible that the reference to the trunk, limbs, branches and twigs of a tree is intended to be analogous to the network structure and hence road pattern.
Although the diagram accompanying Buchanan's tree analogy does not indicate that the network pattern is tree-like over the hierarchy as a whole , it does nothing to discourage the association of minor roads at the lower end of the hierarchy with 'twigs' in the form of cul-de-sac at the extremities of the road network.
Given that the distributors become progressively less connective lower in the hierarchy, it is hard to believe that the access roads are supposed to form a network more connective than the local distributor system (which is fragmented here into discrete, non-overlapping sub-networks), or that access roads would bridge across the cordons sanitarian implied by the blank swathes either side of the district distributors (to which, of course, the access roads are not permitted to connect). Indeed, the deliberate restriction of choice of path between precincts was expressly intended.
Movements between the environmental 'cells' and the interlacing network of distributor roads would be canalized without choice. The conclusion is that the access road - the nearest equivalent to the traditional urban street - was intended to assume a spatially disjointed as well as hierarchically subordinate role in the system.
It seems therefore that the only place for the 'street' (that is, a road with footways and building frontages) in the hierarchy promoted in Traffic in Towns is at the level of the access road that is, a minor road which is not expected to form a contiguous network and which in practice is often characterized as a cul-de-sac .
The cul-de-sac access road is thus both a 'twig' in terms of its position in the road hierarchy and also in the network pattern. In this sense, in the conventional hierarchy, the only role for the street is that of a twig .
However, it is argued here that the role for the street is not that of a twig, either hierarchically or in terms of its position in the network pattern. That is, if the objectives of sustainable mobility and urban quality and vitality are to be realized, then the role of the street cannot be reduced to a disjointed, subordinate access road. In other words, a street is not a twig.
5.1.5 Discussion
The works of Buchanan and Alexander discussed above both appeared within a few years of each other over long years ago. Since then, much has changed - but much has stayed the same. Alexander's work seems to have been absorbed or adopted by much of contemporary urbanism. Despite this, cities are still being conceived and planned and churned out as 'trees'. However, we concentrate on Buchanan's Tree, and follow the issue of road hierarchy.
Since Traffic in Towns, the advocacy for a strict road hierarchy has softened, even within the highways and transport professions. Additionally, it has become more inclusive to some extent, by at least acknowledging the potential role for pedestrian and cycle routes. Transport in the Urban Environment acknowledges that some road categories are 'less clearly distinguished' than others; their roles have to be interpreted flexibly in relation to local circumstances, and proposed hierarchies should be 'open to modification in the light of consultation with local people and affected road-users' .
Nevertheless, advocacy of hierarchy is followed by the observation that 'this idea may be fully achievable only in new development areas'. This appears to indicate that the fully implemented hierarchy is still an ideal, implying that existing urban areas, where hierarchical principles must be compromised, fall short of this ideal. In other words, there is no suggestion that traditional network structures might provide an alternative 'ideal', possessing benefits that the most 'optimal' hierarchy in the conventional sense would struggle to deliver. The presumption is still that, where possible, a hierarchical road layout is best, and the implication is that urban structure will still be dictated by a hierarchy of routes, with streets as subordinate elements, if not actually configurational 'twigs'. It would appear that some sort of synthesis is required, which combines the unambiguous specification of a hierarchy as a typology of routes with explicit roles and relationships with the wider concerns for the resulting forms, and the accommodation of the mixed function urban street. If such a hierarchy is to be developed, it is necessary to investigate not only what types of route are allowable (street, distributor, etc) but what alternative structures are possible for organising these.
5.2 Alternative hierarchical typologies and structures
The previous section has discussed some of the fundamental attributes - and basic problems - of conventional hierarchical structuring. This section looks at a wider range of alternatives, which will pave the way towards reformulating hierarchy. This section firstly briefly considers the possible variety of types (section 5.2.1) which may be assembled into hierarchical typologies (section 5.2.2). These correspond more or less with the first two components (connotations) of the 'tree analogy' discussed in Box 5.1. This is followed by considerations of possible alternative hierarchical structures (section 5.2.3), corresponding broadly with the third component of the 'tree analogy'. Finally, section 5.2.4 considers the relationship between hierarchical structure and network configuration.
5.2.1 Route types
The first component of the 'tree analogy' was the recognition of different types. A wide variety of route types is observed in practice. These types may be directly related to physical or functional characteristics of routes, or may simply be nominal labels.
The pursuit of a more diverse 'hierarchy', to reflect the desired variety of urban route types, suggests the use of a greater number of types, beyond the limited collection of distributor roads and access roads currently admitted. The question becomes one of how the individual labels, such as a selection , may be related to each other in an ordered typology. How can ensure ' Standardisation of use' as well as diversity.
5.2.2 Hierarchical typologies
The second component of the 'tree analogy' was the ordering of types. This section firstly reviews some examples of hierarchical typologies, and then goes on to discuss the possible rationales for their hierarchical ordering.
Table 5.1 shows an illustrative selection of institutional hierarchies - those promoted or applied by public authorities, and featuring in policy guidance. In each case the hierarchy uses some kind of road type as its basis for differentiation. However, while the terminology differs in each case, the basic principles follow the same general pattern:
(1) There is a spectrum from major roads with limited access to minor roads with frontage access, i.e., there tends to be an inverse relationship between traffic and access functions. The result is that roads designated as 'streets', implying built frontages and activities, are normally found to the lower end of the spectrum;
(2) there tends to be greatest segregation of modes implied at either extreme of these hierarchies: segregated vehicular traffic at one end and segregated pedestrians at the other. In fact, in Traffic in Towns, in a rather extreme case, there is segregation between vehicles and pedestrians in all but the 'lowest' level in the hierarchy;
(3) most route types appear to be designated according to transport function, although some also imply relationships with buildings. In only one case, that of the cul-de-sac, is the layout or 'structural role' of the road type explicitly specified, although network characteristics are hinted in terms such as 'trunk collector';
(4) in all but one case, the hierarchy is 'linear' in that there is a simple ranking of road types with only one at each level. Portland's Arterial Streets Classification Policy is an exception in that it has parallel hierarchical designations for auto and transit traffic. We can also observe many non-institutional typologies (Table 5.2).
It's worth noting that there are more types of roads at the lower end of the road hierarchy.
In addition, "street" is used not only at the bottom of the hierarchy. This reflects a conscious desire to reshape traditional urban route types in connection layouts, which is reflected in connection streets. Connecting streets should be able to carry moderate levels of local traffic in a manner compatible with bike and pedestrian traffic. Some people care not only about the road as a pipeline, but also about the city (or rural) streets and space.
Connection street is an example of a road type that may be defined by its structural role in the network. However, the exact limits of this role are not clear. It would be useful to have a clearer resolution of this kind - like a connection network.
So far, we've seen a variety of possible alternative types. The problem now focuses on how to handle the routing type.
Road hierarchy is conventionally regarded as a functional hierarchy, i.e., the roads are classified according to their (intended) function, rather than their form. In fact, the hierarchical classification of route types may serve a range of purposes, for example, to distinguish responsibility for routes, to assist with information, or to distinguish road function or standard. Table 5.3 shows a series of route types and classification criteria. Some of these criteria are based on form type or use type , others on designation or network role.
The correlation between major roads and heavily trafficked roads and strategic roads seems intuitively simple. However, things are not necessarily as straightforward as this in reality. If we look more closely, we found that national networks ultimately tend to be organized on the basis of strategic routes, and not on traffic use. In other words, if such systems are function based, this really means network function, not traffic function (use). The national system of 'A' roads is effectively independent of both form and use; an 'A' road may take the form of a narrow old street in a town, despite its substandard nature as a highway, or a lightly trafficked cross-country route. It is the strategic role of such routes, forming a contiguous network, that is key to their being accorded 'A' road status.
Conversely, a 'hierarchy' based on either traffic use or road form would result in a discontinuous 'mosaic' of route segments across the country. Though in many cases these might indeed link up contiguously, they would not inevitably do so.
The outstanding feature that the national road network does possess is that strategic routes do all connect up in a particular way. This property is described by Morrison as 'arterially', by which the 'pattern of arterial roads is the only one which necessarily forms a complete network' .
There is an eleoe28shVIrdtArG7wduFgXQyvu7yObMRgGEszsjhdJO0=ment of self-referentially about this: roads seem to be defined by how they are connected within the network, yet what they connect to may be limited or ensured by their designated role.
How the different route types may be connected up with each other in a constitutional sense is therefore germane, and shall be considered next.
5.2.3 Hierarchical Structure
At this point the thesis perhaps enters its most abstract section. Here there will be attention to structures at a level of abstraction most removed from what it is the structures are describing: streets and networks.
This section analyses the structure of formal hierarchies, and the way in which different combinations of road type may be arranged in different forms of hierarchical organization. This broadly equates with the third component of the 'tree analogy' in section 5.1.1.
Conventional hierarchy has been criticized for being too rigid, and for not allowing enough road types to be represented. In order to expand or modify conventional hierarchy, it is necessary to explore the structures of hierarchical organization. Although often associated with 'tree-like' layouts (and the occasional use of terms such as 'tree-like' hierarchy), conventional urban road hierarchy may in fact be best regarded as a linear, vertical structure .
This simple diagram can be loaded with two distinct concepts. Firstly, it shows allowable connections between route type. In this example, a route type may only connect to a type immediately adjacent: an access road may connect to a local distributor but not directly to a district distributor.
Secondly, we may also use the vertical ordering of the diagram to indicate arterially: the condition by which all routes down to any level (starting with primary distributors) must form a contiguous system. Hence, primary distributors must form a single network, while pedestrian streets may form a scatter of segments isolated from each other.
Arterially introduces asymmetry into the hierarchy, since it implies different kinds of network connectivity as we move 'up' or 'down' the hierarchy. This demonstrates why, in an abstract constitutional sense, we cannot simply 'invert' the hierarchy, without some changes to network structure.
The fact that there is only one route type at each level, and no alternative connections 'bridging' between nonadjacent levels. In fact, overall, it is possible to consider any theoretical hierarchical structure in terms of its 'dimensionality'; it is possible to see how, with progressive branching in an ordered manner, it would be possible to arrive at a hierarchy in the form of a two-dimensional lattice or matrix of types.
The possibility of a matrix of more than one dimension allows the creation of an alternative formuSY0InJZWgm6cuqjaz5ZfXQ==lation of hierarchy which would allow more flexibility and diversity while retaining some structural organization. For example, a two-dimensional hierarchy could treat pedestrian function, or access function, as an independent variable from traffic function, thereby releasing the necessity for pedestrian or access functions always to be associated with the lowest rung in the (vehicular) hierarchy.
5.2.4 Hierarchical structure and network structure
It is suggested that a linear hierarchy (constitution) is suitable for a conventional tree-like pattern (configuration). A more complex hierarchy would not be particularly useful in this case, since the tree-like pattern implies a narrow range of route types with simple structural connectivity.
However, a more complex hierarchy, such as one possessing a branching or matrix structure, would be able to the potential diversity of route types in traditional grid-like patterns. Linear hierarchies may be applied to traditional grids, as conventionally has been the case (II). The fit is not perfect, however, and leads to the observed narrow definition of all sorts of street types being shoehorned into a single 'access road' category.
The conclusion here is that hierarchical structures more complex than the conventional linear case are possible, and indeed may be more desirable for application to the more complex network structures, with a greater variety of urban route types, as considered desirable by many urban planners and designers. The question then becomes one of putting all these issues together, in a reformulation of hierarchy.
5.3 Reformulating road hierarchy
We have now explored conventional hierarchy, and noted the urban disadvantages of Buchanan's Tree, which in practice has tended to result in a combination of linear hierarchies and tree-like network structures. A means of relating more complex hierarchical structure to more complex network patterns has been suggested. This section suggests a possible new formulation for hierarchy, based on a wider range of types of route, in a multi-dimensional hierarchy, which addresses different modes as well as the built frontage dimension.
Firstly, by use of a wider range of road types it should Q3UmYgFO7g4FFHA4+i/M7EenwuEtn6ZpOPw2wysyU30=be possible to have greater urban variety, greater accommodation of 'existing reality' and the ability to accommodate Street types that mix vehicular traffic function, public transport and pedestrian needs within the immediate context of the built environment - frontages, and access to buildings generally.
Secondly, such an expanded hierarchy allows for more complex patterns to be created. These are not necessarily the 'chaotic' forms which planners originally have wished to do away with, but more structured patterns which represent a kind of 'organized complexity' which allows for variations in use - and not a simplistic division between road-as-traffic-artery and piazza-as-people-place.
In order to create a new formulation for urban road hierarchy, it is necessary to address both the typology - the named road types included - and the hierarchical structure (linear, branching or matrix form). These two items corresponding to the first two components (connotations) of the 'tree analogy' will now be dealt with in the following sections.
5.3.1 Typology
As we have seen, conventional road hierarchy restricts itself to a limited number of road types. At its most extreme, we have seen the number was reduced to two basic types distributor road and access road with the term 'street' effectively disappearing as an officially recognized type.
In reformulating hierarchy, an important consideration is to cater for as broad a range of types to reflect the diverse reality of urban areas, if that diversity is to be maintained and supported.
We earlier saw that the main 'battle' was for the mid range of the hierarchy. Here, indeed, we find suggestions for revising road types. We suggest that "district distributor roads might be designed as high streets that link residential developments together, and onto which houses and a wide range of other uses should face, rather than as fast 'bypasses' lined with lap fencing and backs of houses" . This echoes Russell's advocacy for 'mixed traffic urban street' lying between primary distributors and access roads . Similarly, Allan Jacobs advocates the use of multi-purpose boulevards.
Additionally, and to cater more explicitly for sustainable modes, we may add: . A variety of pedestrian route types, including both pedestrian-only streets and footpaths;
? Public-transport-only routes, such as bus ways and bus gates;
? The 'transit mall', a street used by public transport (bus or tram) and pedestrians, but with no through traffic function (for general traffic).
Some of these types were never encouraged by conventional hierarchy, while others were actively discouraged. Therefore, the existence of a typology that recognizes and encourages these varieties would at least be a start. That said, the typology is no use on its own: it must be accompanied by appropriate hierarchical relationships and backed by appropriate layout structure.
5.3.2 Multi-dimensional hierarchical structure
The problems of conventional road hierarchy may be condensed to the following points:
1. The hierarchy is linear, and ordered by vehicular function;
2. Pedestrian function is pegged to vehicular function, at the lowest rung in the hierarchy;
3. Frontage/access function is pegged to vehicular function, at the lowest rung in the hierarchy;
4. Public transport is not explicitly accommodated. Buses are supposed to use distributor roads: (a) this tends to isolate bus routes from the routes used by pedestrians; (b)the road layout will tend to be geared to the needs of private vehicular traffic.
It is therefore proposed that a potential solution is to replace the linear hierarchy with a multi-dimensional construct that releases pedestrian function, public transport function and frontage function as independent dimensions, that are no longer tied to general traffic function. Then, route types may be based on permutations of those functions, allowing separate types of layout to be tailored to their needs.
We may start by first considering two dimensions: vehicular traffic function (v) and pedestrian function (p).
In practice, not all positions in the matrix are likely to be equally strongly represented by actual road types on the ground. A road with maximum importance for both traffic and pedestrians would nowadays be relatively uncommon (i.e., major traffic flows are normally routed away from major pedestrian flows: the busiest pedestrian routes may often be completely or partially pedestrianised ). Nevertheless, the range of road types in practice is likely to be more than the four or five categories allowed for in the conventional hierarchy, which amounts to being a one-dimensional, traffic-oriented arrangement.
2022e93b283fbdc689d4a9c461f420a5It is then possible to add a third dimension to our hierarchical construct, to represent public transport (or transit) function (t). The presence of a bus, tram or rail service in an urban corridor constitutes an extra dimension which may be regarded, at least theoretically, as being independent of the traffic or pedestrian significance of the corridor. (For simplicity in this discussion we will refer only to the presence of buses, but theoretically the system could include any form of public transport including those partially or completely removed from the public road network.) The addition of the third dimension gives seven permutations of modal presence (Table 5.5).
Finally, we can add a fourth dimension, that of frontage function. This can be used to distinguish between those 'roads' and 'streets' which might have the same modal composition, but which may be markedly different in terms of urban form due to the absence or presence of building frontages.
While traffic function can obviously be described as a continuum, 'access' (that is, to adjacent properties) is more like a yes/no condition. In fact, for simplicity, we could assume that both the third and fourth dimensions are 'either/or' cases, ie, a route either has a public transport service or not, and may either have building frontages or not.
This thesis does not propose hard and fast recommendations for what type of construct ought to be prepared, or which permutations of route type might be the most meaningful: the construct and the types would depend on each other, and both would depend on the circumstances of intended application. However, one possible approach is now suggested, which illustrates one possible selection of route types and hierarchical construct.
5.3.3 The Mixed Mode Hierarchy construct
It is now possible to generate a specific constitutional construct. A set of named types can be equated with different permutations of vehicular, pedestrian, public transport and frontage function. The suggested hierarchical construct will first be described in terms of its constituent layers, from which can be selected a set of types which representing those most likely to be found in practice, or most useful within the context of application.
Firstly, we can consider four layers or 'footplates'. Each layer comprises the 5 by 5 array of cells shown in Figure 5.13, representing permutations of pedestrian and vehicular function. The four layers then represent the different permutations of public transport and frontage function. Let us suppose the following layers:
? footplate I represents routes with neither public transport nor frontage function - among route types on this plate would be conventional distributor roads;
? footplate 2 represents routes with frontage function but no public transport – these would include a range of urban streets;
? footplate 3 represents routes with both frontage function and public transport function - these would represent the major streets served with bus routes;
? footplate 4 represents routes with public transport function but no frontage function - these might include dedicated bus ways or interurban main roads.
It can be seen that the current conventional road hierarchy is still represented within the new formulation, albeit as a special case. To some extent it may be said that the new formulation simply appends extra dimensions - pedestrian, public transport, buildings – to a hierarchy which was simply intended to address the road traffic dimension. However, the new formulation is more properly an elaboration of the conventional hierarchy, which implicitly included the other three dimensions all along, albeit in a fixed relationship with the dominant road traffic function. The new formulation thus releases these other functions as independent dimensions, allowing a wider range of road types to be present, configured into a number of co-present but overlapping hierarchies
5.4 Chapter discussion
This chapter has explored hierarchical structure and made explicit the distinction between constitution (comprising hierarchy and typology) and configuration (topology). This may be added to the previous distinction, between configuration and composition. The distinction between the three may be summarized graphically. Note that all three layout types may be accommodated by the same constitution, hence demonstrating that 'tree-like' layouts need not automatically follow from a 'hierarchical' constitution..
Layouts (a) and (b) have the same tree-like configuration (topology), realized on the ground in different compositions: (a) could represent a curvilinear suburban layout, while (b) could represent an inner city grid with street closures. Layout (c) has a different configuration, that of a connective grid, but has the same constitution as (a) and (b): in each case there is a three-tier hierarchy of routes, where the third tier (thinnest lines) may not connect with the first tier (thickest line).
We have seen how and why hierarchy is viewed negatively by some planners and urban designers. Firstly, it contains a reduced diversity of route type, which has no place for the traditional mixed-use urban street. Secondly, it appears to be geared to traffic function. The principle of ' arterially ' seems to lead to layout structures whose connectivity is geared to the national vehicular network, at the expense of local pedestrian and frontage functions.
The solution proposed here is to treat the different modes, and the presence or absence of building frontages, as independent dimensions. The use of a 'mixed mode' hierarchy can directly assist with a more pedestrian- and public transport-oriented road network. Such a hierarchy could allow the re-creation of traditional street types which could be served by buses or trams, hence bringing together public transport, pedestrians (passenger demand) and buildings (generators of demand).
In a sense this breaks free from the constraints of Buchanan's Tree (the association of hierarchical structure with tree-like layouts) and the implicit idealization of the multi-level segregated, disurban solutions proposed in Traffic in Towns.
The 'mixed mode' hierarchy is no more - but no less - than a formal expression of the reality of contemporary urban road systems, which represent a balance between the severity of Buchanan's ideal and the uncontrolled 'chaos' it was intent on replacing. Within the proposed hierarchical construct, we could explicitly accommodate a ' Tottenham Court Road' and a 'Fitzrovia'. Hence we have a robust means of reconciling neo-traditional aspirations with hierarchy, which has a transparent theoretical grounding, as the new formulation for hierarchy is effectively built up from first principles.
Such a hierarchy is not 'revolutionary', in that it does not involve the complete overthrow of the hierarchical concept, nor does it simply 'invert' the hierarchy, changing the names at the top and the bottom while leaving the basic structure intact. Rather, the approach is evolutionary, in that the proposed hierarchy can almost literally unfold from the existing hierarchy, which is implicitly contained within it.
As it stands, however, the proposed formulation only covers route types and their relative positions in the hierarchical construct . This leaves two unresolved issues.
Firstly, it does not yet specify the allowable - and required - connections between different route types. These issues, relating to access constraint and arterially, shall be returned to, when considering design recommendations in Chapter 10. For the moment, we may assume, at least, that any type may connect with any type in an adjacent cell in the hierarchical table. Secondly, it does not address what types of structure or network configuration would support the hierarchy, in the sense of generating routes whose structural role would fit the labels implied by the hierarchy. That said, it has at least been suggested that a tree-like hierarchy is appropriate for a grid-like network structure; conversely, a grid-like structure may be necessary to generate the desired diversity of route types commensurate with a variety of permutations of pedestrian, vehicular and public transport importance.
Before addressing this correspondence between structure and hierarchy in more detail, it is necessary to gain a clearer understanding of configurational structure. In Chapter 4 it was noted that there was a lack of clear description or characterization of urban structure. This strand is now picked up again, in the following chapter.