Interview with Mike Davis: Part 1

I first discovered Mike Davis's work about a decade ago, through his book City of Quartz, a detailed and poetic look at the social geography of Los Angeles. Perhaps most memorably, City of Quartz describes the militarization of public space in LA, from the impenetrable "panic rooms" of Beverly Hills mansions to the shifting ganglands of South Central. Not only does the Los Angeles Police Department use "a geo-synchronous law enforcement satellite" in their literal oversight of the city, but "thousands of residential rooftops have been painted with identifying street numbers, transforming the aerial view of the city into a huge police grid." In Los Angeles today, "carceral structures have become the new frontier of public architecture."
Many of Davis's conclusions will annoy you – but that's half the point of reading his books.

A more wide-ranging book is Davis's 2002 collection Dead Cities. While it's one of Davis's least cohesive books, it nonetheless ends with an invigorating bang. Its final section, called "Extreme Science," is a perfect example of how Davis's books remain so consistently interesting. We come across asteroid impacts, prehistoric mass extinctions, Victorian disaster fiction, planetary gravitational imbalances, and even the coming regime of human-induced climate change, all in a book ostensibly dedicated to West Coast American urbanism.
Of course, Mike Davis's particular breed of urban sociology has found many detractors – detractors who accuse Davis of falsifying his interviews, performing selective research, deliberately amplifying LA's dark side (whether that means plate tectonics, police brutality, or race riots), and otherwise falling prey to partisan battles in which Davis's classically Marxist approach seems both inadequate and outdated. In fact, these criticisms are all justified in their own ways – yet I still find myself genuinely excited whenever a new book of his hits the bookshop display tables.
In any case, the following interview took place after the publication of Davis's most recent book, Planet of Slums. Having reviewed that book for the Summer 2006 issue of David Haskell's Urban Design Review, I won't dwell on it at length here; but Planet of Slums states its subject matter boldy, on page one. There, Davis writes that we are now at "a watershed in human history, comparable to the Neolithic or Industrial revolutions. For the first time the urban population of the earth will outnumber the rural."
This "urban" population will not find its home inside cities, however, but deep within horrific mega-slums where masked riot police, raw human sewage, toxic metal-plating industries, and emerging diseases all violently co-exist with literally billions of people. Planet of Slums quickly begins to read like some Boschian catalog of our era's most nightmarish consequences. The future, to put it non-judgmentally, will be interesting indeed.
Mike Davis and I spoke via telephone.

BLDGBLOG: First, could you tell me a bit about the actual writing process of Planet of Slums? Was there any travel involved?

Davis: This was almost entirely an armchair journey. What I tried to do was read as much of the current literature on urban poverty, in English, as I could. Having four children, two of them toddlers, I only wish I could visit some of these places. On the other hand, I write from our porch, with a clear view of Tijuana, a city I know fairly well, and that’s influenced a lot of my thinking about these issues – although I tried scrupulously to avoid putting any personal journalism into the narrative.

Really, the book is just an attempt to critically survey and synthesize the literature on global urban poverty, and to expand on this extraordinarily important report of the United Nations – The Challenge of Slums – which came out a few years ago.

BLDGBLOG: So you didn't visit the places you describe?

Davis: Well, I was initially anticipating writing a much longer book, but when I came to what should have been the second half of Planet of Slums – which looks at the politics of the slum – it became just impossible to rely on secondary or specialist literature. I’m now collaborating on a second volume with a young guy named Forrest Hylton, who’s lived for several years in Colombia and Bolivia. I think his first-hand experience and knowledge makes up for most of my deficiencies, and he and I are now producing the second book.

BLDGBLOG: I’m curious about the vocabulary that you use to describe this new “post-urban geography” of global slums: regional corridors, polycentric webs, diffuse urbanism, etc. I’m wondering if you’ve found any consistent forms or structures now arising, as cities turn away from centralized, geographically obvious locations, becoming fractal, slum-like sprawl.

Davis: First of all, the language with which we talk about metropolitan entities and larger-scale urban systems is already eclectic because urban geographers avidly debate these issues. I think there’s little consensus at all about the morphology of what lies beyond the classical city.

The most important debates really arose through discussions of urbanization in southern China, Indonesia, and southeast Asia – and that was about the nature of peri-urbanization on the dynamic periphery of large Third World cities.

BLDGBLOG: And "peri-urbanization" means what?

Davis: It’s where the city and the countryside interpenetrate. The question is: are you, in fact, looking at a snapshot of a very dynamic or perhaps chaotic process? Or will this kind of hybrid quality be preserved over any length of time? These are really open questions.

There are several different discussions here: one on larger-order urban systems – similar to the Atlantic seaboard or Tokyo-Yokohama, where metropolitan areas are linked in continuous physical systems. But then there’s this second debate about the spill-over into the countryside, this new peri-urban reality, where you have very complex mixtures of slums – of poverty – crossed with dumping grounds for people expelled from the center – refugees. Yet amidst all this you have small, middle class enclaves, often new and often gated. You find rural laborers trapped by urban sweatshops, at the same time that urban settlers commute to work in agricultural industries.

This, in a way, is the most interesting – and least-understood – dynamic of global urbanization. As I try to explain in Planet of Slums, peri-urbanism exists in a kind of epistemological fog because it’s not well-studied. The census data and social statistics are notoriously incomplete.

BLDGBLOG: So it's more a question of how to study the slums – who and what to ask, and how to interpret that data? Where to get your funding from?

Davis: At the very least, it’s a challenge of information. Interestingly, this has also become the terrain of a lot of Pentagon thinking about urban warfare. These non-hierarchical, labyrinthine peripheries are what many Pentagon thinkers have fastened onto as one of the most challenging terrains for future wars and other imperial projects. I mean, after a period in which the Pentagon was besotted with trendy management theory – using analogies with Wal-Mart and just-in-time inventory – it now seems to have become obsessed with urban theory – with architecture and city planning. This is happening particularly through things like the RAND Corporation’s Arroyo Center, in Santa Monica.

The U.S. has such an extraordinary ability to destroy hierarchical urban systems, to take out centralized urban structures, but it has had no success in the Sadr Cities of the world.

BLDGBLOG: I don't know – they leveled Fallujah, using tank-mounted bulldozers and Daisy Cutter bombs –

Davis: But the city was soon re-inhabited by the same insurgents they tried to force out. I think the slum is universally recognized by military planners today as a challenge. I wouldn’t be at all surprised if there’s a great leap forward in our understanding of what’s happening on the peripheries of Third World cities because of the needs of Pentagon strategists and local military planners. For instance, Andean anthropology made a big leap forward in the 1960s and early 1970s when Che Guevara and his guerilla fighters became a problem.

I think there’s a consensus, both on the left and the right, that it’s the slum peripheries of poor Third World cities that have become a decisive geopolitical space. That space is now a military challenge – as much as it is an epistemological challenge, both for sociologists and for military planners.

BLDGBLOG: What kind of imaginative role do you see slums playing today? On the one hand, there's a kind of CIA-inspired vision of irrational anti-Americanism, mere breeding grounds for terrorism; on the other, you find books like The Constant Gardener, in which the Third World poor are portrayed as innocent, naive, and totally unthreatening, patiently awaiting their liberal salvation. Whose imaginination is it in which these fantasies play out?

Davis: I think, actually, that if Blade Runner was once the imaginative icon of our urban future, then the Blade Runner of this generation is Black Hawk Down – a movie I must admit I’m drawn to to see again and again. Just the choreography of it – the staging of it – is stunning. But I think that film really is the cinematic icon for this new frontier of civilization: the “white man’s burden” of the urban slum and its videogame-like menacing armies, with their RPGs in hand, battling heroic techno-warriors and Delta Force Army Rangers. It’s a profound military fantasy. I don’t think any movie since The Sands of Iwo Jima has enlisted more kids in the Marines than Black Hawk Down. In a moral sense, of course, it’s a terrifying film, because it's an arcade game – and who could possibly count all the Somalis that are killed?

BLDGBLOG: It’s even filmed like a first-person shooter. Several times you're actually watching from right behind the gun.

Davis: It’s by Ridley Scott, isn’t it?

BLDGBLOG: Yeah – which is interesting, because he also directed Blade Runner.

Davis: Exactly. And he did Black Rain, didn’t he?

BLDGBLOG: The cryptic threat of late-1980s Japan…

Davis: Ridley Scott – more than anyone in Hollywood – has really defined the alien Other.

Of course, in reality, it’s not white guys in the Rangers who make up most of the military presence overseas: it’s mostly slum kids themselves, from American inner cities. The new imperialism – like the old imperialism – has this advantage, that the metropolis itself is so violent, with such concentrated poverty, that it produces excellent warriors for these far-flung military campaigns. I remember reading a brilliant book once by a former professor of mine, at the University of Edinburgh, on British imperial warfare in the nineteenth century. He showed, against every expectation, that, in fact, most often for the British Army, in imperial wars, what was decisive wasn’t their possession of better weapons, or artillery, or Maxim guns: it was the ability of the British soldier to engage in personal carnage, hand-to-hand combat, up close with bayonets – and that was strictly a function of the brutality of life in British slums.

Now, if you read the literature on warfare today, this is what the Pentagon’s really capitalizing on: they’re using the American inner city as a kind of combat laboratory, in addition to these urban test ranges they’ve built to study their new technologies. The slum dwellers’ response to this, and it’s a response that has yet to be answered – and maybe it’s unanswerable – is the poor man’s Air Force: the car bomb. That’s the subject of another book I’m finishing up right now, a short history of the car bomb. That has to be one of the most decisive military innovations of the late twentieth century. If you look at what’s happening in Iraq, it may be the Improvised Explosive Devices (IEDs) that are killing Americans, but what’s just ripping that country apart is these fortified car bomb attacks. The car bomb has given poor people in slums – small groups and networks – a new, extremely traumatic kind of geopolitical leverage.

What’s happened, I think, at the end of the 20th century – and at the beginning of the 21st – is that the outcasts have discovered these extraordinarily cheap and horrific weapons. That's why I argue, in Planet of Slums, that they have “the gods of chaos” on their side.

BLDGBLOG: Beyond a turn toward violence and insurgency, do you see any intentional, organized systems of self-government emerging in the slums? Is there a slum “mayor,” for instance, or a kind of slum city hall? In other words, who would a non-military power negotiate with in the first place?

Davis: Organization in the slums is, of course, extraordinarily diverse. The subject of the second book – that I’m writing with Forrest Hylton – will be what kinds of trends and unities exist within that diversity. Because in the same city – for instance, in a large Latin American city – you’ll find everything from Pentacostal churches to the Sendero Luminoso, to reformist organizations and neoliberal NGOs. Over very short periods of time there are rapid swings in popularity from one to the other – and back. It’s very difficult to find a directionality in that, or to predict where things might go.

But what is clear, over the last decade, is that the poor – and not just the poor in classical urban neighborhoods, but the poor who, for a long time, have been organized in leftwing parties, or religious groups, or populist parties – this new poor, on the fringes of the city, have been organizing themselves massively over the last decade. You have to be struck by both the number and the political importance of some of these emerging movements, whether that’s Sadr, in Iraq, or an equivalent slum-based social movement in Buenos Aires. Clearly, in the last decade, there have been dramatic increases in the organization of the urban poor, who are making new and, in some cases, unprecedented demands for political and economic participation. And where they are totally excluded, they make their voices heard in other ways.

BLDGBLOG: Like using car bombs?

Davis: I mean taking steps toward formal democracy. Because the other part of your question concerns the politics of poor cities. I’m sure that somebody could write a book arguing that one of the great developments of the last ten or fifteen years has been increased democratization in many cities. For instance, in cities that did not have consolidated governments, or where mayors were appointed by a central administration, you now have elections, and elected mayors – like in Mexico City.

What’s so striking, in almost all of these cases, is that even where there’s increased formal democracy – where more people are voting – those votes actually have little consequence. That’s for two reasons: one is because the fiscal systems of big cities in the Third World are, with few exceptions, so regressive and corrupt, with so few resources, that it’s almost impossible to redistribute those resources to voting people. The second reason is that, in so many cities – India is a great example of this – when you have more populist or participatory elections, the real power is simply transferred into executive agencies, industrial authorities, and development authorities of all kinds, which tend to be local vehicles for World Bank investment. Those agencies are almost entirely out of the control of the local people. They may even be appointed by the state or by a provisional – sometimes national – government.

This means that the democratic path to control over cities – and, above all, control over resources for urban reform – remains incredibly elusive in most places.

(This interview continues in Part Two. For another, recent two-part interview with Mike Davis, see TomDispatch: Part 1, Part 2. All drawings used in this interview are by Leah Beeferman, who was also behind BLDGBLOG's Helicopter Archipelago).

Comments are moderated.

If it's not spam, it will appear here shortly!

Anonymous Anonymous said...

For the most part, I agree with what is being said but sometimes theory requires a bit of fact checking. While I am not using this to dispute the over-arching argument, I do get tired of the post-modernist myth of the Iraq War being a 'minority-led/minority sacrifice for the white power structure' war tossed around.

from your interview:

"Of course, in reality, it’s not white guys in the Rangers who make up most of the military presence overseas: it’s mostly slum kids themselves, from American inner cities."

The following are two excerpts:

But a close examination of Pentagon statistics suggests that at least some of the conventional wisdom about who is most at risk during wartime is misleading. For example, although blacks account for 26% of Army troops, they make up a much smaller percentage of those in front-line combat units, the most likely to be killed or injured in a conventional war.

In all four military branches, black recruits tend to favor support jobs, from mechanic to unit administrator, over traditional combat slots such as infantryman or fighter pilot. That division of labor is well known within the armed forces.



The report shows there are some significant differences in the makeup of the military and the civilian work force in the U.S.

• Whites are underrepresented in the military. The U.S. work force is 71 percent Caucasian — or other ethnic groups included as whites — while the military is 67 percent white.

• Blacks are overrepresented, comprising 17 percent of the military and 11 percent of the civilian work force

• Hispanics are underrepresented, making up 9 percent of the military and 11 percent of the work force.

• Women, not surprisingly, are also underrepresented, comprising just 16 percent of the military, although policies and laws restricting assignment possibilities for women are the chief reason, not other recruiting factors.

• Deaths of U.S. troops in Operations Iraqi Freedom and Enduring Freedom also have been disproportionate to ethnic makeup of the military, the report states. Based on the 1,841 deaths and 12,658 wounded service members as of May 28, when the report was being prepared, 71 percent of the dead are classified as white, 9 percent as black and 10 percent as Hispanic.


May 23, 2006 7:10 AM  
Blogger e-tat said...

Really interesting stuff. It's good to see an interview of some substance, and the detailed response by Davis. Among the things I am provoked by, the one about peri-urbanization prompts a quick remark. I don't know what L.A. is like, but London, Boston, Amsterdam and who knows where else have redefined their peripheries in such as way that they cannot be regarded as countryside. They are sub-urban, but not to be confused with the suburbs like Welwyn Garden City or Braintree. And, of course, these first world places could hardly be the urban entities described by Davis. The point then, is to avoid thinking in terms of the old dichotomy of rus/urb, and find another way of defining these spaces as they become a distinct feature of human social organisation in relation to a landscape.

May 23, 2006 7:31 AM  
Anonymous Anonymous said...

I'm with anonymous. While the interview itself is very interesting, the military figures don't add up and they don't back up his politics.

May 23, 2006 8:50 AM  
Anonymous Anonymous said...

Good points from the anonymice, but nonetheless, the phrase, "they’re [the Pentagon] using the American inner city as a kind of combat laboratory," is a spur to the most entertaining conspiracy theories. For example, they aren't resurfacing any of the roads in Center City Phila. after digging them up a month ago, because "they" are surreptiously training urban drivers to deal with potholed and object-strewn roads. Mayor Street's Neighborhood Transformation Initiative is obviously a Pentagon-directed strategic terrain change, from dense yet crumbling housing stock to a situtation that provides a great deal less cover--like Fallujah post-daisy-cutters-and-bull-dozers ...

Anyway,I'm 85% positive we can't infer the Pentagon's current preferred/planned mode of combat just from a trip through North Philly, but speculating about urban administration and planning failures in this way does make a nice change from frustration.

Great interview, by the way--I'm looking forward to the next installment.

May 23, 2006 10:56 AM  
Blogger Michiel said...

If we are into conspiracy theories: Geoff, is your blog also sponsored by the Ministry of Defense, the Pentagon? ;-)

May 24, 2006 8:48 AM  
Blogger jpb said...

I enjoyed seeing this interview, not only because I like Davis a lot but also because I'm currently about halfway through City of Slums, and this interview has served nicely as a kind of "commentary track" running alongside it. Thanks!

May 24, 2006 1:33 PM  
Blogger Geoff Manaugh said...

jpb, glad the interview's useful. And Michiel, of course - the NSA underwrites this whole thing. And now we've got your IP address... Living in Holland is no way to escape.

e-tat, I think that peri-urbanism is exactly the problem you describe; in other words, peri-urbanism already is the recognition that there is a spatial and socio-geographic complication that goes unaddressed by terms traditionally used to discuss 21st-century urbanism. In other words, again, Davis is saying - or Davis seems to be saying - that peri-urbanism is already this topologically complex mixing and overlapping of different spatial types and strategies of land use; so the term already recognizes the non-rural aspects of the rural, as well as anti-urban pockets that exist with the urban. I do think that the concept of a "perimeter" brings with it a lot of unnecessary baggage, including the idea that there are easily-defined limits, but that's a weakness inherent to the term, not inherent to Davis's analysis. Giving him the benefit of the doubt, of course.

My point, then, is that peri-urbanism is already not a dichotomy between rural and urban.

Nicky: Philadelphia as a street-hazard driving school. There's an architectural project in that somewhere...

And, anon, thanks for the figures and links.

May 25, 2006 1:19 PM  
Blogger Geoff Manaugh said...

Violence intensifies across Mogadishu (BBC).

May 25, 2006 2:52 PM  
Blogger e-tat said...

Geoff, what you're saying makes sense at least insofar as the rural has never been neatly separate from the urban, given that industrialising cities initially drew their workforce from the sticks, and in some instances (such as the UK) cities were collection basins for people who were being forced off the land. There is also the phenomena of ruralites bringing their country ways to the city, whether it's goats on the balconies in Luanda or grazing on the old railway embankment off Brick Lane in London, and not simply limited to a bit of livestock, but the effort to map village life onto the urban fabric. In this sense, the urban is a means of preserving some aspects of the rural, and the two are never neatly separated.

My point is that even this is not a good framework for thinking about urban cosmographies. Rather, I may be thinking more along the lines of something touched on in the interview, which is that the Victorian idea of cities - which worked for those kinds of urbanisations - cannot recognise the new urban ecologies. The word 'city' could become obsolete in some cases. The idea of a discrete entity with a centrally-administered management might give way to something more fluid, ad hoc, more extensive, and less comprehensible. The big question would then be about who stands to benefit, and how, from such modes of organisation.

May 26, 2006 2:17 PM  
Blogger Geoff Manaugh said...

e-tat, I totally agree with you. My point was just that the word "peri-urban" is already meant to describe such a situation - a situation in which, as you write, "a discrete entity with a centrally-administered management might give way to something more fluid, ad hoc, more extensive, and less comprehensible." That is peri-urbanism.

May 26, 2006 2:55 PM  
Blogger e-tat said...

It is? Well howcome I'm the last one to find out that's what it is? Heck, I've never even been to one of those places! I suppose I should read the book, or something.

May 26, 2006 5:18 PM  
Blogger e-tat said...

Hold that thought. I've got a question. If we think of the organisation of urban spaces as a matter of spatial ideology, where an idea about space is adopted/promoted/developed by an influential group of people, then it's pertinent to ask who is doing the imagining.

If, for example, cities have largely been imagined and orchestrated via town planners, but that in cities where planners are overwhelmed by circumstance, the imaginings are done by squatters, immigrants, gangs and slumlords, then one ideology develops formally through government, think-tanks, academic institutions, and journalism while the other develops informally through some other means. Some of these means are regressive. Just look at the kinds of things Robert Neuwirth reports. But there may be some that are indicative of new aspirational ideas of urban life - in ways that are captured by Pentecostalism and so on.

So the question is, is that what Davis is writing about?

May 27, 2006 6:49 AM  
Anonymous matt said...

Thanks for this really fascinating interview. I am glad to see that hear, unlike in the book, Davis speaks more about the mobilisation in the slums. Check out, for instance, Abahlali baseMjondolo - the inspiring shack dwellers' movement in South Africa.

April 06, 2010 2:11 PM  

Post a Comment