Monday, March 24, 2014

Dissolving Border Vacuums, Part 9

Institutional campuses sometimes induce the same border vacuums along their perimeters as those found around parks and “green spaces.” These vacuums are perhaps surprising because institutions like hospitals and universities are vital urban amenities. So I hasten to remind readers that even though I'll be critiquing the urban form of several noteworthy Baltimore institutions in this post, this should not be taken as criticism of the value of these institutions!

As discussed in the previous post, parks can sink into dereliction if they're not activated with “cross-use.” As Jacobs argues in Chapter 14, the city side of a park perimeter needs to offer an enclosure that feeds people into the park, and the park side of the perimeter needs to offer edge programming to lure those people into the park.

If the park is poorly enclosed (i.e. it lacks sufficient mixed-use infill), if its programming is concentrated in a secluded center, or if the boundary between the park and enclosing neighborhoods is too unpleasant and/or difficult to cross, then the cross-use doesn't occur and a border vacuum emerges.

The same pattern holds true for hospital, university, and other institutional campuses: if there is no cross-use between the campus and the enclosing urban fabric, then a border vacuum emerges. I think the vast “dead zone” around Johns Hopkins Hospital in East Baltimore is one of the most depressing examples of this phenomenon. Many rowhouse neighborhoods adjacent to the country's finest medical institution languish in terrible decay: there clearly has been little cross-use between these neighborhoods and Johns Hopkins Hospital!

St. Hedwig’s Hospital in Berlin supports lively mixed-use neighborhoods.
This outcome is all the more disappointing when we contrast the Johns Hopkins Hospital border vacuum with the dynamic surroundings of great medical institutions in other cities: usually the neighborhoods around these institutions are some of the liveliest and most productive in their respective cities, filled with the students, working professionals, and ancillary businesses that support the institutions. In short, there is ample cross-use between the institutions and their surrounding neighborhoods, something that is sorely lacking around Johns Hopkins Hospital.

To its credit, Johns Hopkins has initiated several projects to overcome the massive border vacuum surrounding its East Baltimore campus. A substantial amount of infill housing for students and young professionals has been built adjacent to the campus, and the hospital has injected resources into surrounding educational and commercial anchors.

A view of the massive “scorched earth” border vacuum north of Hopkins Hospital.
However, I think some of these projects have relied on unfortunate “one step forward, two steps back” redevelopment tactics. For example, the “scorched earth” tactic of acquiring block after block of rowhouses, condemning and clearing them en masse, and replacing them with equally-vast swathes of uniform infill runs the risk of swapping out one iteration of failed “gray area” for a second iteration of “gray area.”

As discussed in the parking lots post, infill needs to feature a small increment of development to be affordable, accessible, and practical for a sufficiently-broad range of residents and businesses. As Jacobs warns us in Chapter 10, rapidly-built-up districts with a limited range of building types tend to incorporate the seeds of their future decay.

Perhaps this is why Johns Hopkins felt it had to resort to such large-scale clearance in the first place: the original rowhouse districts surrounding the hospital were themselves built up too rapidly with too limited a range of building types (almost nothing other than two-bay, two-story rowhouses), and these “gray area” districts never transitioned into the diverse, mixed-use, mixed-form neighborhoods necessary for supporting a diverse medical institution. There was, for example, almost no commerce outside the Monument Street corridor. Like many other struggling, single-use, “gray area” rowhouse districts in Baltimore, this district was arguably destined for malaise from its inception in the 19th century.

An example of recent infill construction near Johns Hopkins Hospital.
But will urban-renewal-style clearance solve the problem? The replacement housing does incorporate excellent urban design (it's largely built out to the sidewalk without resorting to blank walls), but again, I worry that the increment of development is too unwieldy, and that the range of infill types/forms is too narrow.

I think a better redevelopment approach would have (1) engaged a wider range of developers to produce a wider range of infill buildings, (2) urged the city to undertake comprehensive zoning deregulation in the district to allow a richer array of bottom-up property owners and entrepreneurs to repurpose/replace the existing building stock on a finer grain, and (3) downsized/subdivided the district's blocks via laneways and alleyways (something the old rowhouse blocks actually did better than the new infill) to avoid superblock formation, with its attendant tendency to undermine cross-use.

For example, there's been an encouraging amount of rowhouse renovation along upper Broadway, but the fundamental 19th century shortcoming of this stretch of Broadway – that it was built up with a single building type for a single use – remains and should be overcome. Zoning deregulation and incremental conversion/replacement of some of the rowhouses with other building types would, I think, gradually allow the mixed-use vibrancy of lower Broadway in Fell's Point to spread northward.

Another example of the “one step forward, two steps back” redevelopment approach around Hopkins Hospital is the striking contrast between the form of the infill housing and the form of infill hospital buildings. As discussed earlier, for all the shortcomings in development scale, the infill housing does follow good urban design principles. Unfortunately this isn't the case with many of the campus' medical facilities: there is a heavy reliance on superblocks composed of parking lots, parking garages, and blank-walled buildings connected to each other via skywalks and concourses.

How Hopkins Hospital interacted with surrounding rowhouse neighborhoods.
I think this further undermines the viability of the adjacent infill housing: why bother living next to a blank-walled superblock if you don't get any benefits (such as street life, “eyes on the street” safety, convenient commerce, and ease of cross-use) from the adjacency? This is why the old rowhouse fabric decayed as badly as it did: it had no value for reuse because it had no practical connection to the campus.

This is hardly a criticism of institutional agglomeration or even parking per se: institutions like hospitals quite understandably need to concentrate their facilities to accommodate internal cross-use, and skywalks and concourses easily accomplish this. Furthermore, institutions like Hopkins Hospital, which receives a substantial number of national and international patients, need to offer some parking. I merely think amenities intended to foster internal cross-use become problematic when they begin to undermine external cross-use between the institution and its surrounding neighborhoods, and this is clearly the problem around Hopkins Hospital.

For example, despite their usefulness, I think skywalks unintentionally reinforce the border vacuums they're crossing in a destructive feedback loop: the bridges that connect the Orleans Street Garage to the Bloomberg Children's Center ultimately allow Orleans Street to devolve into an ever-nastier traffic sewer, which only makes it more difficult for the vibrancy along lower Broadway to leap across Orleans and spread northward.

Fortunately there are solutions for institutional superblocks, and Johns Hopkins Hospital seems to be increasingly receptive to them:

Firstly, if it is necessary to close a street to enlarge a block, the resulting superblock can retain some porosity by offering promenades that connect to surrounding streets. A continuous street grid is thus preserved for pedestrians and bicyclists, if not for automobiles. Of course, as Jacobs warns us in Chapter 9, promenades are pointless if there's no reason for people to use them – they need to contain the same enclosing mixed uses that ordinary streets do. Fortunately the two promenades in the western half of the Hopkins medical campus – the Jefferson and McElderry Promenades, which were once conventional streets – offer this, in addition to regular event programming such as farmers' markets.

Jefferson Square at Washington Hill places its parking in the center of the block.
Secondly, rather than placing their blank walls, parking garages, and unpleasant-but-necessary service infrastructure (loading docks, ducts and grilles, service doors, etc.) along the street, institutional facilities can employ the “perimeter block” format to conceal these activities in the centers of blocks. Jefferson Square at Washington Hill does this, though it missed an opportunity to extend Fairmount Avenue across its superblock. Parking garages can also incorporate retail liners (as at Johns Hopkins' Rutland Garage), though there needs to be a sufficient pool of residents nearby to support them.

Thirdly, rather than concentrating their semipublic programming deep in the interiors of buildings, institutional facilities can move these activities into retail liners that open directly onto surrounding streets. A quick perusal of the Johns Hopkins Hospital map reveals that the campus' internal loop offers numerous ATMs, chapels, coffee shops, flower shops, gift shops, cafes, restaurants, food courts, lounges, and pharmacies for visitors. But what if this internal loop fronted onto surrounding streets to enliven them and offer amenities for visitors and nearby residents? Perhaps this loop could even have been designed as a fully-public series of midblock commercial laneways that seamlessly flowed into surrounding neighborhoods, potentially preventing their decades-long slide into irrelevance in the first place!

It's perhaps too late to employ these tactics within the existing campus (though there are opportunities for thoughtful retrofitting), but I'd encourage the hospital to consider them as it expands into surrounding blocks.

From many conversations I've gotten the impression that we often assume the best way to experience a perfect “walkable” neighborhood is to visit a medieval town in Europe. But many of us live in similar “walkable” places – college campuses – for several impressionable years in our youth and then forget this when we move into the “real world!”

I've often wondered why this disconnect between our collective experience of a “walkable” college campus and the “real world” exists, and I think the conventional City Beautiful campus is itself to blame. For many of us they are indeed our first and only experience of “walkable” communities, but many also tend to be introverted to a degree such that we may be subconsciously concluding they are “special” places – wonderful to experience for a few years, but irrelevant to the structure and operation of the “real world” urban fabric: “Typically they pretend to be cloistered or countrified places, nostalgically denying their transplantation, or else they pretend to be office buildings (267).”

This sense of separation may also be responsible for the “town-gown” friction between universities and their host cities. The friction is also a result of the hypertrophic “eds and meds” bubble of recent decades, in which universities and hospitals have rapidly expanded into surrounding neighborhoods. Many large state universities, for example, have induced vast “student ghettos” on their peripheries, with all their attendant friction.

The Hopkins Homewood campus is incredibly beautiful, but it sits in a bubble.
In Baltimore, the campus fortress mentality can be observed in the so-called “JHU fear bubble” around Johns Hopkins' Homewood campus. Hopkins is attempting to overcome this insularity by expanding the range of mixed-use, mixed-form buildings in Charles Village (the rowhouse neighborhood adjacent to the Homewood campus), essentially employing Jacobs' “extraordinarily strong counterforces” vacuum-fighting tactic.

In addition to the incremental densification of Charles Village, I think Hopkins could “activate the edge” of its Homewood campus by moving more of the campus' semipublic programming closer to adjacent streets, as Jacobs suggests: “Universities could make their campuses more like seams and less like barriers if they placed their uses intended for the public at strategic points on their perimeters (267).”

For example, future student centers, theaters and performance spaces, clubhouses, galleries, and even dormitories could be placed along the Homewood campus' Charles Street and University Parkway frontages (the Lacrosse Hall of Fame is a useful precedent in this regard). Wyman Park could also be transformed from an overgrown back door “wilderness” into a vibrant front door between the Homewood campus and Hampden.

These campus perimeter tactics, useful as they are, eventually raise the question: do urban universities even need campuses? If they were part and parcel of the ordinary urban fabric, would it be easier for us to make the leap from “walkable” university neighborhoods to “real world” neighborhoods equally hospitable to “walkable” working, shopping, entertaining, and child-rearing? I think the campusless university is better poised to accommodate this mental leap – and perhaps is already doing so in younger generations.

What exactly are “campusless” universities? These are essentially universities that were established in a single building (or perhaps in a small collection of buildings organized around a modest courtyard or quad), but have since grown into prominent institutions by acquiring disparate buildings across their host cities, rather than decamping to blank slates on the fringes of those cities to establish discrete campuses. The University City district in Philadelphia is a famous example, as is the Savannah College of Art and Design, and Baltimore has the local examples of MICA, the University of Maryland, and the University of Baltimore.

The University of Wroclaw is seamlessly integrated into its host city.
Such universities are even more common abroad. Some – such as the Jagiellonian University in Krakow and the University of Wroclaw – are so beautifully integrated into their host cities that there are no discernible campus edges and no consequent border vacuums. Rather than bestowing a sense of importance on their buildings via “open space” setbacks, these universities use lavish architecture to distinguish their buildings from surrounding background buildings.

Furthermore, instead of being cloistered on a single site (or even in adjacent spillover “student ghettos”), campusless students are generally distributed across various neighborhoods around a city. There are fewer “turf lines” between campus and city, and students consequently mature earlier – they learn to use their host city for daily life, rather than consciously leaving special campuses for special trips “into the city.”

How the University of Wroclaw interacts with its host city.
More specifically, campusless students quickly learn how to use the public transit, restaurants, and other amenities in their host cities, rather than having to rely on the infantilizing substitutes – shuttles and dining halls – that conventional campuses provide.

I therefore think campusless students have a better opportunity to transfer their urban educational experience to an equally-viable urban “real world” experience after graduation: “I slept, ate out, did groceries, work-studied, ran errands, explored, and commuted to classes on foot, bike, and transit across this city during my college years, and I can easily continue doing this after I graduate.”

How a generic university campus typically interacts with its host city.
A conventional campus student, on the other hand, arguably has a harder time transferring their bubble experience to a mismatched “real world.” Many cozy campus “villages” don't really teach us how to interact with their greater urban organisms, so we forget their “walkable” settings surprisingly quickly after graduation: if your experience of “walkable” living is confined to a campus, how can you transfer that special experience once you have to leave the campus?

Many campusless universities still have courtyards and quads, but these spaces often serve students and city residents, since they're so thoroughly integrated into the urban fabric. For example, it'd be difficult for a university to claim a plaza like this as its own “turf,” since so many city residents casually use it for shortcuts, meanderings, and relaxation.

I think such fine-grained cross-use helps ease “town-gown” friction, since it becomes harder to divide people into oppositional camps – there's too much sharing of public amenities, and too much mutual benefit. Many conventional City Beautiful campuses, on the other hand, often use their poorly-enclosed “open space” grounds to clearly demarcate their “turf,” with little benefit for adjacent residents and little consequent cross-use.

This isn't to say that campusless universities have impeccable “town-gown” relationships: as observable in places like Philadelphia's University City, there still is friction if the universities are undergoing hypertrophic growth. Blank-walled starchitecture can likewise undermine the connection between university buildings and the supporting background fabric. But campusless universities still seem to be better at easing “town-gown” friction than discrete campuses experiencing a sudden “bursting of the floodgates.”

I would hardly argue that universities like Johns Hopkins should dismantle their existing City Beautiful campuses – in these cases I think the previously-discussed edge activation strategy is preferable. Not all university campuses are problematic either – many college towns are enlivened by their presence, precisely because these campuses take great pains to activate their edges with cross-use. But when it comes to establishing new universities or expanding existing ones, I think the campusless approach is preferable.

Midtown-Belvedere enjoys considerable cross-use from MICA and UB.
Fortunately American universities increasingly seem to be adopting this approach – fewer and fewer are doing what Goucher College did in the 1950s when it decamped from central Baltimore to a suburban campus. Many are instead acquiring disparate buildings across their host cities, ensuring the fine-grained integration of students and city residents, thus preventing the formation of border vacuum turfs.

For all the incredible beauty of a detached Homewood campus, I'd rather have a lively Midtown/Mount Vernon enjoying thorough cross-use from MICA/UB students and city residents alike.

Next time we'll wrap up the series on border vacuums with Jacobs' thoughts on vacuum-inducing neighborhood “turfs” and the “great blight of dullness” (a blight afflicting large swathes of Baltimore!), so stay tuned!

- Marc Szarkowski


  1. Marc, while this is an excellent article I would like to clarify some of your Philadelphia examples.

    (1) UPenn is not a "campusless" university. (On the contrary, it has an exceptionally large campus for an urban university.) Rather, the reason University City succeeds so well is because the school laid excellent seams between the campus and the city along 40th and Walnut Sts.

    (2) Drexel University is planning on taking this a step further; it intends to convert the majority of its campus' acreage into a gigantic seam and, in fact, construct an entire neighborhood (possibly two or three) on what is currently campus and campus-adjacent grounds.

    The better example in Philadelphia would be the University of the Arts, a major arts school occupying several buildings along the Avenue of the Arts. No discernable campus exists; in fact, school functions are tightly interwoven amongst condos, theaters, retail, offices, etc.

    (3) The only other major American university that I am aware of, that even comes close to being truly campusless, is New York's New School.

  2. Hi Steve,

    Point taken on UPenn! You're absolutely right regarding the seams; the first time I wandered into the district I was impressed with how well UPenn extended nearby streets right through the campus. (I.e. Woodland Ave/Walk, Locust Street/Walk, 36th/37th/39th Streets, not to mention the numerous conventional auto/bus/trolley streets passing through the campus.) So I suppose I mistook it for a campusless district because the sense of crossing a "turf line" between the campus and adjoining neighborhoods is nonexistent in many places.

    The campus isn't perfect (the Harrison/Rodin Houses resort to tower-in-the-park layouts, and not a few buildings present blank or otherwise desolate frontages to the street), but it seems UPenn is adopting the campusless approach as it expands to the west and northwest. Perhaps this goes back a bit as well: I wonder if the Evans Building, for example, was at one time almost completely surrounded by rowhouses, and only after WWII did the rest of the campus expand west to meet it (judging by the difference in architecture).

    In actuality I think there's a spectrum between "campusless" and "conventional campus," and that most urban universities fall somewhere in between. It's also possible for a prewar "campusless" university to undergo hypertrophic growth and acquire nearly every surrounding building, in effect transforming itself into a campus overlaid on the urban street-and-block chassis. This can work well if there are lively seams, as you mentioned, though in the US at least the accompanying "student ghettos" always seem to generate friction.


Related Posts Plugin for WordPress, Blogger...