City of Sound is about cities, design, architecture, music, media, politics and more. Written by Dan Hill since 2001.

Los Angeles: Grand Theft Reality

Written in


Los Angeles plays itself, via films and videogames

A portmanteau post loosely arranged around Los Angeles, starting with a few words on the Michael Mann film Collateral. It’s an enjoyable film, in retrospect; no classic, perhaps. However, it’s set in LA—in fact, more than simply “set in LA” but essentially made out of LA, which is reason enough to see the film.

It’s beautifully shot, on high-definition digital video cameras (Thomson Grass Valley Viper FilmStream and Sony CineAlta.) The screen fizzes with neon signs and streetlights, burnt orange skies polluted with light and smog, reflections raytraced from mirrored corporate towers, spots of blue liquid crystal glow from mobile phones lending theatrical uplighting on characters’ faces. Sight and Sound described it, thus:

Probably the first studio director to embrace digital for its purely aesthetic potential, Mann uses the high-definition technology — in particular its ability to register a rich array of colours and tones in low light and at night — to realise his vision of the city. If the LA of Heat was crisp, almost photorealist in its high-gloss intensity, here the night-time cityscape is rendered with a watercolour delicacy. Collateral’s after-hours timescale may be classic noir, but Mann’s subtle night-vision — the dark sky seems bathed in an urban glow (the pale wash of orange street lights, streaks of white automobile headlights) — softens the genre’s chiaroscuro tendencies. [“It Happened One Night”, Sight & Sound, on Collateral]

Mann, in interview:

(Y)ou can’t see the city at night on motion picture film the way you can on digital video … I think this is the first serious major motion picture done in digital video that is photoreal, rather than using it for effects. DV is also a more painterly medium: you can see what you’ve done as you shoot … Digital isn’t a medium for directors who aren’t interested in visualisation …[“Paint It Black”, Sight & Sound]

Crop from ‘Collateral’, Michael Mann (dir.), 2004

Yet again, Los Angeles is a ‘visualised’ city. Twas ever thus, which leaves such interesting traces on our understanding of the city — and not just visually, but formally, as noted in Mike Davis’s City of Quartz, perhaps a peerless book on LA, at least with respect to both that imagined city of Raymond Chandler and Chinatown, and various other murky political and social movements throughout the 20th century. Reading City of Quartz tends to mean that one subsequently engages with almost any film set in LA in terms of a far richer back-history of imagined and real corruption, beauty, horror, sensual image and vivid music, exploitation, aspiration, politics, crime, grime and so on. It’s as if each film is simply a small scene in a growing ‘meta-work’ about the city (more on this later).

With Grand Theft Auto: San Andreas just emerged and, well, extraordinarily good, Collateral is also a reminder of how far video games still have to travel in order to visually catch up with auteurs like Mann. Whilst Mann’s no doubt right about DV enabling photoreal, the results still look hyperreal to me, or “painterly” in results as well as in process. When video games stop attempting to mirror normal movies and start looking like the extraordinary, exaggerated, hyperrealism that Mann is achieving, then we’re going places. Literally.

To be fair, I think San Andreas does begin to steer in this direction. The sheer visual flair with which it imagines LA is one of the more impressive features of this episode in the series. The game itself feels no more than a gentle extrapolation of Vice City and a change of scene — and nothing wrong with that really — but the graphics of the city at night, all dreamy neon haze blurred by speed, are quite lovely. Equally, observe the cold morning light filling the windshield briefly before that low trajectory of the sun lifts up to generate the shimmering heat of midday. The various zones of the city — bearing in mind I’ve explored less than 10% of the enormous map thus far — are beginning to reveal their identities, formless opaque blocks becoming distinct characters.

Screenshot from ‘Grand Theft Auto: San Andreas’ (2004)

One common complaint about the game is the small and indistinct on-screen map, though I have to say that the only real way to play GTA is to drive around endlessly, building your own mental map of the city. To me, this is just as in reality. Only tourists navigate using a map every few minutes. Games obviously shouldn’t replicate all of life’s travails — part of a game is to escape from all that — but this one aspect of forcing ‘imageability’, as Kevin Lynch described it, seems fundamental to imagining a city. Building a coherent image of the city is fundamental to GTA, and that only comes from slowly recognising all the landmarks, paths, networks, edges, longitudinal streets, cross streets, and so on. Here’s Lynch describing how one builds up images of different districts of late-50s Boston:

(T)he relative width of the streets, the block lengths, the building frontages, the naming system, the relative length and number of the two kinds of streets, their functional importance, all tend to reinforce this differentiation. Thus a regular pattern is given form and character. [From Kevin Lynch’s The Image of the City [Amazon UK|US]

The same experiential approach is required in GTA, and the fictional city feels all the more real for relying on you to conjure it into existence, form and character inferred from seeing the fabric of the city, rather than being pinned down by an abstraction, a map. In this, there are echoes of Lynch’s theories in Sight and Sound’s review of Collateral, navigating the city via landmarks and distinct neighbourhoods:

In Mann’s portrait of the megalopolis, the beaches, bikinis and eternal sunshine of celluloid California are replaced by a tenuously interconnected set of zones containing black neighbourhoods, Korean nightclubs and Latino gangsters. Our journey back and forth across the city is so well mapped — with the glass spires of downtown providing a point of orientation — that no matter how incredible the plotting, one’s sense of location is always spot on. [“It Happened One Night”, Sight & Sound]

This detailed sense of location is common to Grand Theft Auto too, generated from painstaking research. According to an article in Blueprint, game creators Rockstar start with a “long, 80-person research trip, and carries on throughout the project with more trips, as the team refines new locations needed within the maps, such as nightclubs, garages and restaurants.”

Despite lifting this aspect of imaging real cities, creating detailed districts from street furniture to typography to sound design, GTA’s cities are deliberately not real, relying on a form of ‘closure’ (cf. Scott McCloud) to construct the city. Rockstar’s Dan Hauser says:

We could a real city in the game, but we chose not to. We create an approximation, an abbreviation of a real city, carefully designed to support all the visual and typographical variety we want. It’s more like a 3D set than an actual city. This games is about creating a reinterpretation of the US, a socially and visually distorted prism of the real thing … Experience has taught everybody that it’s better to make something that looks good, seems real, and carries a punchline in what it says. We try to make a world that, at first glance, seems completely normal, and then reveals its absurdity as you play it. It looks better, and works better within the medium. [“Game of Life”, Blueprint, December 2004]

This call-and-response, shuttling back and forth between fiction and reality is immensely powerful, creating enjoyably unsettling experiences. For example, GTA’s fictional LA recently pranged my recent memories of real LA in an almighty high-speed mental pile-up.

I was exploring a bit of beachfront in the game, named Santa Maria, idly driving around (in between missions, I could take my time). And I had this sudden, vivid sense of deja vu. I realised they’d recreated the Santa Monica beach almost exactly, which I’d briefly visited a couple of months earlier. I wandered around, in character, retracing my steps. Up on to the pier, on to the beach, along the curious mixture of expensive postmodernist homes and tatty old huts facing the beach. It was a very odd feeling indeed. It wasn’t exact — the pedestrian bridge spanning the freeway wasn’t there, disappointingly given its claustrophobic feel, nor were the cafes on the pier present. The lifeguard’s huts were a bit abstracted, nor were the beachfront houses exact analogues. But they were accurate gestures towards the houses that are actually there. And given the experiential nature of this imagined city, the gestural approach, requiring a certain amount of closure from the player, generated a far more powerful feeling of being in two places at once and yet in neither.

Take a look at these photos of me in the game, in ‘Santa Maria’, hastily snapped from off the telly — apologies for the quality — and compare with photos I took of the real Santa Monica, a couple of months earlier. Admittedly, without the similar construction of memories and experiences around these still images, you may not feel the same spooky resonances I did.

Perhaps this is even an example of where the real world encroached perhaps just a little too closely, counter to the gestural approach described above. Rockstar’s art director Aaron Garbut:

Converting a real city into a game is restricting. As soon as you label something ‘real world’, people’s perception of it changes. I don’t think it’s possible to be completely immersed in a game if you’re constantly making comparisons. [“Game of Life”, Blueprint, December 2004]

As mentioned previously, the fictional cities in GTA are a riot of architectural fakes and extrapolated reality. I’ve really only explored Los Santos (Los Angeles) thus far, but I’m still getting a real sense of the Los Angeles I briefly discovered over a couple of days there earlier this year. It’s a truism barely worth stating that Los Angeles is a city constructed from images entwined with fake histories, and as noted, the best book exploring this terrain would be City of Quartz (see also Celluloid Skyline, by James Sanders, on New York). And the best film would appear to not be Collateral or Chinatown or Heat, but Thom Anderson’s Los Angeles Plays Itself, a collage-like movie constructed of hundreds of unlicensed clips of LA on film, with a central premise outlined in this early part of the voiceover:

If we can appreciate documentaries for their dramatic qualities, perhaps we can appreciate fictional films for their documentary revelations.

Subject of a fantastic piece in last week’s Guardian by John Patterson, it’s showing at the ICA in London till mid-January 2005. Do go and see it if you can, as it’s absolutely brilliant. I’d love to get hold of the script, as there were some very wise things said about cities and the way we construct them. A three-hour (almost) collage of film clips ‘about’ Los Angeles— little of it licensed, I imagine—taking in popular film, radical cinema, black cinema, noir, SF etc., with a smart, witty, provocative, occasionally cranky, and wide-ranging voiceover. Almost like Godard’s Histoire du Cinema, though not as densely layered — it’s more sequential collage, rather than montage—and focused on a single city; in fact, a film about a city formally constructed out of film of the city itself.

Finally on Los Angeles, I’m yet to find the doppelganger of Frank Gehry’s Walt Disney Concert Hall in Grand Theft Auto: San Andreas. Wouldn’t it be a delicious irony that the computer-generated buildings that Gehry’s studio knocks out are the most difficult to render in computer-generated faux cities like Los Santos? Perhaps this why it isn’t there. The buildings in GTA are often complex, but not that complex. I managed to fire off a few photos of Gehry’s building, as the cold blue LA night began to drape downtown’s glowing skyscrapers.

This, after an odd ‘Englishman-in-Los-Angeles’ afternoon of getting a bus from Santa Monica — yes, a bus, LA residents! The 50 dollar taxi fare downtown seemed a bit hefty, whereas heading for downtown, the bus was quick, cheap and a good way to see the city. I was the only non-Latino/non-African American on the bus for most of the journey, until a couple of German tourists got on. I then got lost as I had no idea where to get off, and ended up out past Beverly Hills and West Hollywood, walking — yes, walking, LA residents! — for some time back towards town. At this point, I realised that LA was a very large and distinctly non-European city in urban form, despite walking through an area which seemed to be briefly 100% Russian in population. Having ploughed on through areas entirely without pavement, I ended up getting a taxi towards downtown, driven by a hugely entertaining and warm Iranian emigre cabbie (“So, how is my friend Mr Tony Blair?”, “Er…”) at a cost of 40 bucks. I then spent 15 minutes forcing the Walt Disney Concert Hall to twist and pose in response to my direction (like David Hemmings in Blow Up, with Gehry’s building played by the equally glacial and unresponsive Vanessa Redgrave), before heading back to Santa Monica with my new Iranian/American friend, at the cost of another 40 bucks.

Gehry’s building, as with the MIT Stata, seemed fabulous, complex, beautiful, and problematic in terms of integration with the surrounding environment. But this being downtown LA, integration with the surrounding environment is essentially a problematic notion whatever the building. But I didn’t really have time to do anything other than fire off these photos. I couldn’t find a way to sneak in, as a concert seemed about to start and LA’s finest were stepping out of taxis and taking in a brief 1-minute inhalation of the outside world before entering another hermetically-sealed culturally-gated space.

Last words to Michael Mann, who starts with an odd first sentence, given Anderson’s Los Angeles Plays Itself, but goes on to make a fascinating point about the topography of LA, as compared with the internet. I saw what he meant when I was on that bus, getting lost in LA. Most people don’t see LA. As Thom Anderson says: “Who knows Los Angeles best? … The people who walk. People who take the bus.” The 39 out of 40 LA residents who don’t work in the movies. There is a real Los Angeles buried underneath the weight of celluloid, and now digital, history. How ironic that Michael Mann, a filmmaker obsessed with surfaces, points it out:

Los Angeles is relatively undiscovered as a location for pictures. It has its own pattern of culture which is almost like travelling on the internet — when you visit websites you journey through some kind of interstitial space and the domains are not contiguous. I think LA is the most exciting contemporary city in the United States, but you have to know where to go, and most Angelinos don’t.

This piece originally published at on December 20, 2004. Photos from a subsequent trip, five years later.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: