What am I missing about New Urbanism's Transect?

(Please read this post carefully and to the end before commenting. Thanks.)

•••

Summary added after writing the post:

1. I agree with the Transect.
2. I agree with Form-Based Coding.
3. I am bewildered why so many people assert (but without explaining) that there is an essential theoretical, didactic or practical link between the two.
4. This post is not an attack on New Urbanism.

•••

I received static on what someone took to be my dismissal here of the the Transect. Of course I didn't intend to dismiss the substance of the Transect at all and in fact allowed that it might well be accurate. I was  only trying to poke a bit at its jargony Star-Trekky name; I hadn't really thought about the Transect enough to offer  an opinion one way or another about its substance, so I didn't.

A  correspondent, however, made the reasonable suggestion that  I look into the Transect. So I did. And I hold to my sentiment about the name, which is not particularly important. But my conclusion as to the substance is now bewilderment.

Here's a description of  the Transect:

A transect of nature is a geographical  cross-section of a region intended to reveal a  sequence of environments. It helps study the many symbiotic elements that contribute to habitats where certain plants and animals thrive.  

Human beings also thrive in different habitats. Some would never choose to live in an urban core, and some would wither in a rural place.  To provide meaningful choices in living arrangements, the rural-to-urban Transect is divided into six T-zones for application on zoning maps. These six habitats vary by the ratio and level of intensity of their natural, built, and social components. They are coordinated by these T-zones to all scales of planning, from the region through the community scale down to the   individual lot and building.

And it goes on, somewhat portentously:

A transect was first used for biogeographical analysis by naturalist Alexander von Humboldt in the late 18th Century. In the late 20th century in Miami Beach, Andrés Duany and his brother Douglas identified a rural-to-urban transect of the built environment from the T-1 beach through T-3 and T-4 neighborhood fabric to T-5 and T-6 mixed use corridors.

The Transect says that there is gradient in the intensity of human settlement, all the way from the nomad's tent to the Burj Dubai. Of course that is true and obvious. How can one disagree? Are we really expected to believe that no one had ever before noticed that there is a gradient of intensity in human development? The word sub-urban itself suggests such a gradient. The geographer von Thunen wrote about central places and rent gradients in the first half of the nineteenth century. I would bet that there are 18th-19th century novelists or diarists, going back to the Romans, who describe in detail a journey from urban hustle-bustle to wind-on-the-lonely-beach.  (Poets, too? Think Wordsworth?) The gradient of intensity of human activity from, say, Times Square to Mount Marcy is self-apparent. Most importantly, our current Euclidean zoning codes already reflect the understanding of that gradient (or transect, if you prefer) though of course crudely and with the wrong result by my standards. Even the most sprawl-inducing contemporary land use codes express the idea that cities have centers and that human activity decreases in intensity outward from those centers.

I can see the criticism that post-WW2 planning simplified the transect and removed whole levels and that the form of these centers are uncivil etc etc  One might also argue that there is a good and bad profile to transects and that a good city needs every level. But it seems to me that setting up such a large system to get across such a basic point (there is gradient in the intensity of human settlement) is a waste of energy and diverts attention from the task at hand.

But the man behind the Transect, Andrés Duany, has proven himself to be an innovative thinker. (I have no doubt that urban historians will see him — along with Elizabeth Plater-Zyberk and a very few others — as a seminal figure in the urbanism of our era, of far longer-lasting importance than any starchitect now working.) So obviously I might well be missing something big. Someone please help me out. Surely, for example, there must be a fair-handed discussion of The Transect out there? Critical but coming from basically inside the tent?

So call me a skeptic but not unpersuadable.

•••

Btw, the definition of the Transect above is taken from SmartCode Central and I want to separate my bewilderment about the Transect from criticism of "form based codes" as described at SmartCode Central. See also Form-Based Code. (Btw, that last site doesn't even mention The Transect anywhere so maybe others agree that it is not essential.)

I don't know enough about "form-based codes" yet to opine, but my preliminary sense is that "form-based codes" are exactly the right way to go. And that should be no surprise as The Three Rules is itself is an extremely abstracted (but to the point) form-based "codelet." Form-based codes shift the focus to a desired end-state rather than merely offer a set of procedures which have far less-clear design goal.  So of course I suspect I will be in sympathy with the larger and more complete versions found at such places as Form-Based Code and SmartCode Central.

•••

Here's a nice slide show on the subject with the title For those who don't understand the Transect.... Lots of lovely pictures but the narration is missing so one can't hear John Massengale's explanation.

Comments

[Bfdonnelly]

I'd like to help, but I'm not sure what you're missing about the Transect. It's not that nobody thought of a natural-to-urban gradient before, but rather that it wasn't worked into a code before.

I think it's a little bit of myth-making to date it from Miami Beach. I'm sure there was an "aha" moment, but the idea has an earlier lineage. The lots at Seaside were organized into different "types" on which people were to build different types of buildings. A Type I lot had to have a Type I building and so on. Already back then the "types" were organized into a gradient -- though not yet a Transect -- from modest Florida Crackers to mixed-use buildings. (Seaside was once semi-affordable.) The one-to-one correspondence between lot types and building types missed the variety of building types found in good old urbanism, yet the gradient from more rural to more urban was definitely helpful. Therefore, broader categories -- the T-zones -- were organized. Each T-zone could host a variety of building types, types of frontage, types of landscaping, and so on. The Transect zones let you build a townhouse, say, in more than one T-zone -- T4 or T5 -- and the also allow more than one building type into each T-zone. In T5, you can have townhouses, apartment buildings, small office buildings and more. So the Transect is a more flexible system than the original highly controlling idea of a 1:1 correspondence between the location of a lot and the buildings on it.

The Transect takes maybe 5 minutes to explain with the right graphics. Its main purpose, though, is to provide a flexible system for allocating building types, types of frontage, landscaping and even light poles so that they are consistent with each other. The Transect system is just as complex as it has to be in order to give people the flexibility of having more than one type of buildng (etc.) on a block without its all falling into a hodgepodge.

Have you seen the SmartCode? You can download a copy at www.smartcodecentral.com. You can go to Table 14 and see how things are allocated to the T-zones.

If you're not doing coding or design, though, it may not be all that relevant if what you need is a quick-and-dirty answer and not a whole system that is meant for everyplace and every type of good urbanism.

[Sandy Sorlien]

Hello Dave - I'm the editor of the model SmartCode and co-author (with Andres Duany and others) of the Manual for the code. Discussions about the Transect are almost constant on various New Urbanist listservs, including two SmartCode lists comprised of scores of people who have customized transect-based codes for many years. There is plenty of internal critique. It's essential to have it, because SmartCodes and other transect-based codes are in demand and we need to solve numerous widespread problems ranging from suburban retrofit to urban infill to regional structure, not to mention the green stampede of one-size-fits-all strategies that threaten to over-green the city and degrade the very qualities that make cities dense and walkable.

I can't write much at the moment, because I'm wiped out from several days of manning the Center for Applied Transect Studies booth at the EcoCity conference in San Francisco this week. Regarding the arcane nature of the Transect (that name is basically internal shorthand for the community scale rural-to-urban transect we use for zoning reform), I've spent these days explaining it to numerous EcoCity delegates who wander up to the booth and say, "What is this transect all about?" Invariably they get it in two to five minutes. It's actually one of those "elegant" theories scientists are always seeking.

I promise I will get back to this blog with more information for you, but I didn't want to wait too long before saying something.

In the meantime, we bought your book and know what your work is about, how about buying ours? Link on the SmartCode Central home page.

Cheers
Sandy

[Will Cox]

If one is spending so much time explaining the concept, it is probably detracting from one's ultimate goal.

[Sandy Sorlien]

David asked about the connection between form-based codes and the Transect. (Note that the latter term is internal shorthand for the rural-to-urban transect of the built environment.) The SmartCode and other transect-based codes are form-based codes, but as you have noted, not all form-based codes are transect-based. The connection is this: Each transect zone is made up of typological elements (buildings, thoroughfares, frontages, plantings, etc.) that are defined mainly by their form, not their use. That makes it a form-based code. It can show form by way of diagrams, but the presence or absence of diagrams isn't really the determinant. The SmartCode without diagrams is still form-based.

You make an interesting point about conventional zoning being based on a rudimentary rural-to-urban transect. That's true, but the problem is that, unlike an ecological transect where the interdependence of plants and animals is studied, conventional codes don't do a good job of identifying and coding for the different elements that make up each zone and create its immersive character. Use and density simply aren't the only components of a habitat that contribute to its character. For example, conventional codes pay almost no attention to private frontages, which are crucial to walkability, eyes on the street, natural cooling, and retail success. You need (and find) different frontages in different habitats (T-zones).

We've used the term "simplexity" to describe the transect methodology - a simple way to allow and encourage complexity of urbanism. The Three Rules have potential, but they are simply too simple; they allow some bad stuff. I guess I'll comment on that on that thread.

Sandy

[john massengale]

"Are we really expected to believe that no one had ever before noticed that there is a gradient of intensity in human development?"

No, which is why you won't find anyone making that claim. :-) But if you look at the last 50 years of building in the US, you'll find an awful lot of planners, architects and developers building with very little attention to location and context.

There's a free-standing McDonald's in a Manhattan parking lot that's the same as a few hundred suburban, small town and rural McDonald's. There are virtually identical glass towers in downtown Oshkosh, by gosh, as exurban Dallas Fort Worth. Etc., etc., etc.

Sandy et al know I agree with your criticism of some of the language around the SmartCode and the New Urban Lexicon. Usually I'm told that the specific example cited is the right one for its audience. The example you've cited seems to be aimed at engineers and planners, as a lot of SmartCode language is.

[Sandy Sorlien]

Dave asked me to elaborate on my comment that the Three Rules would allow some bad stuff. I just re-read the chapter, and agree that the Three Rules (if amended) are useful for some contexts. However, there needs to be an introduction stating that the Three Rules are for downtown shopping streets of towns and cities only, i.e., what is usually T5 and T6 on the rural-to-urban transect. At the moment, it just says they are for "cities," as if a city neighborhood were nothing but shopping/mixed-use streets. Most of the best urbanism in the world has residential areas adjacent to the shopping streets - they support the retail after all, and provide varying habitats for residents - and much of that has setbacks for front gardens, dooryards, porches, stoops, etc. (A residence should not be at grade on the sidewalk - no privacy.) So Rule Number One must respond to context. (Note- the transect methodology helps solve this.)

Once the shopping street rule is established, I would amend the rule about parking to this:
"Parking, if inevitable, should be behind, above, or below buildings, never in front (except onstreet) or beside. Same goes for the secondary frontages of corner buildings."

If you don't talk about corners, you will get parking lots at the corner and a very weak block face. You need walkable frontages to go around corners into the "Elm Street" areas so that people will walk from their homes to the shopping street. (Note- the transect methodology helps solve this.)

By the same token, we shouldn't encourage lots beside the buildings. That creates the "missing teeth" syndrome that makes for a discontinuous walking/shopping environment. I just returned from San Francisco where I saw virtually zero surface parking lots on frontages downtown or in two other neighborhoods. It was quite startling and refreshing. That's probably simply because the land values are so high, compared to my city of Philadelphia where they are not. Their code may also prevent them, as a good code should.

Since David has to write several pages of text to explain the Three Rules and all the exceptions to them, maybe they aren't so simple after all. But if you do want to make some simple rules for T5 streets, you could amend the Three Rules as above, but add two more:

4. "The block face of a shopping street should be continuous, with only minimal breaks for parking lot entrances."

5. "Buildings on shopping streets should be at least two stories."

This last one accomplishes several things. It provides for apartments over shops, helping with affordable housing (business owners can live over shop, or rent them out) and a safe 24-hr presence on the block. It reduces overall parking requirements by enabling shared parking spaces. And it helps create the feeling of the "outdoor room" in the public street space.

Cheers,
Sandy

[Sandy Sorlien]

John Massengale wrote:
"Sandy et al know I agree with your criticism of some of the language around the SmartCode and the New Urban Lexicon. Usually I'm told that the specific example cited is the right one for its audience. The example you've cited seems to be aimed at engineers and planners, as a lot of SmartCode language is."

It's true that some SmartCode language is meant to dovetail with existing planner/engineer language while re-framing it for the forces of good. ;) However, the terminology and passages David cited are aimed at environmentalists, who use transects for the analysis of natural habitats. When you show anyone (even an engineer!) a diagram of a natural transect (which most people recognize from their high-school textbooks) and then a diagram of a transect extended into the built environment, they get it right away.

Environmentalists need to understand that humans deserve healthy habitats too, and that we must plan natural and urban habitats together to solve the great problems we face. It's starting to happen; I was heartened by the recent EcoCity conference in San Francisco, where urban density was a huge positive topic.

Sandy