The Death of Cinema

Aphoristic, cryptic and apocalyptic, this is a frustrating, infuriating little book. So naturally I based my whole PhD on it. When I say ‘based’, I mean more in terms of inspiration than actual analysis or theoretical underpinning, for no matter how many times I returned to it, there was always something at the centre of Cherchi Usai’s book that seemed to utterly elude me, and I could never write about it with the kind of controlled and comprehending, all-embracing perspective that I thought - rightly or wrongly - was expected of me. The genre of the academic dissertation doesn’t lend itself to the approach that film critic Francesco Cassetti took when he presented an earlier version of this book at the Cineteca Italiana in Milan in 1999 and proceeded to tear off its pages one by one - an ingenious solution to the problem of how to make sense of Cherchi Usai’s work.

At the time when he came to New Zealand to deliver the keynote presentation at the Film and History conference held at our national Film Archive back in 2000, Paolo Cherchi Usai was senior curator of the motion picture department at George Eastman House and director of the L. Jeffrey Selznick School of Film Preservation - impressive credentials indeed for a commentator who might have otherwise exposed himself to the charge of simply not understanding the potential of digital technologies in the area of film storage, transmission and long-term preservation. And since I was finding myself similarly reluctant, albeit in a broader and less expert domain, to accept the dominant ideology of information and networked computing, his example from the coalface provided much needed support and inspiration.

The day of that conference in Wellington, Cherchi Usai didn’t claim specifically that cinema was dead but made a point that would have been even more provocative to his audience of historians and film historians, namely that the moving image isn’t a document. Both of these counter-intuitive notions - that cinema is unsuited to document the empirical real, and that besides it’s dead anyway - are based on Cherchi Usai’s unique understanding of the moving image, its history and the circumstances that enable its production and consumption. That last term in particular, consumption, is crucial here for its double meaning of fruition and destruction, and its capacity to convey therefore the peculiar being in time of the moving image, which is progressively and inexorably degraded each time we look at it.



Images in decay are the subject of Bill Morrison and Michael Gordon’s extraordinary Decasia, a symphony on the fleeting nature of cinema. But understanding that materials like nitrate and acetate don’t last forever is all too easy, and can be represented just as easily; the digital is something else entirely.

When information loses its body and becomes a pure mathematical abstraction, how do you even fantasise about its loss? We know what a corrupted digital image looks like. We can even simulate it by applying glitch effects or using tools such as Digieffects Damage. But a much more common circumstance in which this decay manifests itself is through a file that just won't open, or a DVD that stubbornly refuses to play. Suddenly there is no longer anything to see. As Cherchi Usai documents in his book, when Pixar went to make the DVD of the original Toy Story, it discovered that

twelve percent of the digital masters had already vanished. For three months, Pixar Animation Studios staff scoured the system for the toys' missing parts - salvaging all but one percent of what had been lost in the computers. The remaining scenes were reassembled. For subsequent Pixar movies, Lasseter said, 'we have a better backup system'. (p. 100)

You’ve got to backup your data, man. Who hasn’t learned this lesson the hard way? And we should, really, for lots of copies keep stuff safe. But just how many copies, and just how much stuff? As Cherchi Usai told his Wellington audience, it was estimated at the time that 1.5 billion hours of moving images had been produced in 1999 alone. I don’t have updated estimates, but I suspect that the projection he proposed back then of 3 billion hours for the year 2010 would be conservative - that was before YouTube and video-enabled cellular phones. Even at that conservative rate, the amount would reach 100 billion hours by the year 2023 or 2024, and once we’re faced with that order of magnitude, will it really matter how much of it is meant for instantaneous consumption - or no consumption at all - rather than in the hope of repeated viewings and a prolonged archival existence? I suspect not, for the sheer weight of the former would likely overwhelm the latter anyhow, and the task of assigning relevance and context to individual items would be of a level of difficulty that stretches the imagination. In fact, if one buys into the myth that digital information can be copied perfectly and stored and retrieved indefinitely, the prospect are even more grim, Library of Babel-style. Writes Cherchi Usai:

If all moving images were available, the massive fact of their presence would impede any effort to establish criteria of relevance - more so, indeed, than if they had all been obliterated, for then, at least, selective comprehension would be replaced by pure conjecture. (p. 19)

Limit cases rightly invite scepticism, and the author's contention that some time this century we'll reach a singularity followed by a collapse of our systems for storing and exchanging audiovisual content is perhaps the easiest of the imaginings on offer to dismiss, if only because it's so hard to imagine: what would such a rupture look like? But even if you don't buy this side of the argument, how about the events that are not at all hard to imagine: global wars, economic crises, companies filing for bankruptcy even at times of relative prosperity, the no longer profitable trendsetters of last month making way for new ventures. We all know what's happening this coming October 26th, right? And how do you feel about Twitter leveraging its users to 'create value' and leaving the issue of how to figure out a revenue stream for later? How comfortable are you that once that phantom business model has taken actual shape the free will remain free, or that those servers one day won't be bought out and gutted to make space for something else? What will happen then to the world's conversation, and to the illusive idea of the commons that it sustains?

The "guillotine", a machine for shredding 35mm feature films after their commercial distribution, located in Cinisello Balsamo near Milan, Italy. (Image from The Death of Cinema, p. 76.)

Similar commercial imperatives regulate the life and death of moving images, and are even stricter in the film industry proper. In Due dollari al chilo (Two dollars a kilo), Paolo Lipari has documented the activities of two Italian facilities where polyester and triacetate film are recycled in order to produce combs, benches and low-cost fuel for other plants. Says the owner of one of these facilities (my translation):

The films that go through our plant suffer a dramatic change in value. Titanic, which cost hundreds of millions to make and grossed even more, arrived here on seven trailer trucks. When it left, on the same seven trucks, it was worth 25 or 30 thousand dollars. Two dollars a kilo. It's a crude way of considering what we do here. About 40 dollars for each film. 40 dollars for a film that affected us, enriched us, entertained us.

But here's the twist: 'Cinema,' contends Cherch Usai, 'is the art of destroying moving images' (p. 7), and therefore death becomes it and is in fact integral to it. The model image immune from decay that the proponents of the digital seductively invite us to imagine doesn't exist, and if it did, it would have no history (p. 41, at least in my interpretation). It follows, in one of the most inspired passages of the book, that film history should be primarily 'concerned with mutations affecting the physical evidence of the image rather than with any evolution in the expression of meanings' (p. 83). And moreover

[m]oving image preservation will then be redefined as the science of its gradual loss and the art of coping with the consequences, very much like a physician who has accepted the inevitability of death even while he continues to fight for the patient's life. In monitoring the progress of image decay, the conservator assumes the responsibility of following the process until the image has vanished altogether, or ensures its migration to another kind of visual experience, while interpreting the meaning of the loss for the benefit of future generations. (p. 105)

For the reasons stated at the outset, I hesitate to sum up the book in anything resembling a declarative sentence, but the concepts expressed in the two paragraphs above are to my mind the key to a cultural project of immense value, and that goes beyond cinema alone. It's a project that recognises and embraces the transitory nature of our cultural artefacts; that values knowledge over information and context over content alone; that treats our collective and individual capacity for attention and comprehension as finite resources to be carefully managed; and finally that takes a longer view in its guardianship, striving to set systems in place that might allow our cultural heritage to remain alive with connections not in the perpetual present of the digital archive, but rather in a concrete future of material realities that we cannot hope to foresee. In the most unlikely of ways, it is a very hopeful project.

I shall return to this time and again, as you've probably come to expect of me by now. But before I go I want to share with you another example of the kind of intervention - in the spirit of Mr. Cassetti's page-ripping rampage - that this peculiar little book can inspire in its readers. Earlier this year I lent The Death of Cinema to the redoubtable Steven Crawford, and when he returned it I discovered he had fashioned for it a most fitting bookmark, this one:



Brilliant.



Paolo Cherchi Usai. The Death of Cinema - History, Cultural Memory and the Digital Dark Age. London: bfi, 2001.
Paolo Cherchi Usai. Keynote presentation at the Film and History Conference held at the Film Archive in Wellington, 30 November 2000.

The Dullest Person I've Ever Come Across

I’m not particularly interesting.

Gordon Bell

I was at the National Library in Wellington this time two weeks ago for a forum on the National Digital Heritage Archive, a project that I’m going to write about in some detail in the coming weeks and months. It was a very enjoyable occasion, and the library needs to be commended for opening it to the public free of charge, as well as for putting the presentation materials online. I’m not going to discuss any of that today, though, I just want to share with you a phrase that Penny Carnaby used in her introduction to the day's proceedings:

We live in a ‘press delete’ generation.

It’s a point I heard her make before, and it’s always struck me because it’s the exact opposite of how I’d characterise our default setting in relation to the digital products of our culture.

How to preserve born-digital objects is of course an important issue in these relatively early stage of the age of bits, and in her capacity of chief librarian of our National Library Ms Carnaby has a right as well as the duty to be concerned. But it seems to me that what is being lost for the most part is not being actively deleted, but vanishes rather due to lack of care, both cultural (digital objects need to circulate in order to remain viable) and technical (to combat the obsolescence of hardware and software, as well as the physical decay of the storage media). I also happen to think, not unrelatedly, that the act of pressing delete has in fact a lot going for it.

For there is another threat our efforts of preservation, and it’s the rate of content generation. I’m waiting for the next HMI report from the Global Information Industry Center to tell us how much information there might be in the world, and just exactly how staggeringly fast it might be growing. As of 2003, the year of the last report, the estimate included the following take-home points:

Print, film, magnetic, and optical storage media produced about 5 exabytes of new information in 2002. Ninety-two percent of the new information was stored on magnetic media, mostly in hard disks.

and

We estimate that the amount of new information stored on paper, film, magnetic, and optical media has about doubled in the last three years.

An exabyte equals one quintillion bytes, which is to say ten to the eighteenth power. It really is impossible to contemplate a figure like that other than by going ‘wow, that’s loads’. And besides I have all sorts of problems with measuring the information in the world in bytes, as if it were all digital - but I’ll leave that for another rant. The more meaningful and human-readable conclusion is in fact the second one: the amount of new information annually produced doubled between 1999 and 2002. Now ask yourself: has this situation changed in the last six or seven years? Are the cultural and technological conditions in which new information gets produced and stored substantially different, in a way that is likely to have slowed down this rate of increase? If anything, I would say it’s probably the opposite. Blogging, Twitter and Flickr alone would have made sure of that.

So we could equally be saying: we live in a ‘press save' generation, for Save is in fact the default behaviour, when in any sort of doubt, of a computer system in relation to a digital object. Want evidence? When you instruct a computer to delete a file, you are generally asked if you’re sure you want to do it, but when you go to save it, whirr, it just happens.

If you equate information with knowledge, as so many people do nowadays, you may of course opine that none of this constitutes a problem. I will come back to this in a future instalment, and take issue with astute commentator Manuel Castells on whether the talk of an information glut is legitimate and what some of the cultural ramifications of that kind of thinking might be. But today I want to be less subtle and deal with a limit case that perhaps we can all agree upon.

Meet Gordon Bell, IT luminary, former Microsoft executive and dedicated lifelogger. I didn’t seek permission to use the friendly-looking picture on his home page, so kindly navigate to the link above for a glimpse of how he rolls. You’ll see he’s carrying two devices around his neck: one appears to be a voice recorder, while the other is a SenseCam, an experimental Microsoft camera that operates in permanent stand-by mode and wakes up to take pictures every time it sense the heat likely to be emanated by a human body transiting in front of Mr Bell or a change in the lighting conditions suggesting that its carrier might have moved to a new location. This is all part of MyLifeBits, a project initiated by Bell in 1998 but whose origins could be traced back not so much to the visionary genius of Vannevar Bush - as he maintains - but rather to something that Bell’s employer wrote in 1996 in The Road Ahead:

[T]he highway will also make it possible for an individual to keep track of his or her own whereabouts--to lead what we might call "a documented life." Your wallet PC will be able to keep audio, time, location, and eventually even video records of everything that happens to you. It will be able to record every word you say and every word said to you, as well as body temperature, blood pressure, barometric pressure, and a variety of other data about you and your surroundings. It will be able to track your interactions with the highway--all of the commands you issue, the messages you send, and the people you call or who call you. The resulting record will be the ultimate diary and autobiography, if you want one.^[1]

Bell wants one, and he’s doing just that. It started with a lot of old-fashioned digitisin’ - of books, scrapbooks, collectibles, papers - and is continuing nowadays with the recording of every waking moment of Bell’s life. As of May 2007, when The New Yorker wrote a piece on MyLifeBits, the archive included, besides the non-born digital objects

a hundred and twenty-two thousand e-mails; fifty-eight thousand photographs; thousands of recordings of phone calls he has made; every Web page he has visited and instant-messaging exchange he has conducted since 2003; all the activity of his desktop (which windows, for example, he has opened); eight hundred pages of health records, including information on the life of the battery in his pacemaker; and a sprawling category he describes as “ephemera,” which contains such things as books he has written and books from his library; the labels of bottles of wine he has enjoyed; and the record of a bicycle trip through Burgundy, where he tried to eat in as many starred restaurants as he could (he averaged 2.2 stars per meal—“I do a lot of measuring,” he says).

There is a flash in the last sentence of my beloved Georges Perec, who amongst other self-archival activities at the very border of what others would call sanity attempted an ‘Inventory of the Liquid and Solid Foodstuffs Ingurgitated by [Him] in the Course of the Year Nineteen Hundred and Seventy-Four’^[2], but without any of the poetic self-examinations and reflections that accompanied those endeavours. I saw Bell speak at Victoria University in May of 2006, in front of an audience of information technology students, and not once did he pause to consider the staggering ethical and philosophical implications of the project. Come question time, it was all that he was asked about, and he proceeded to politely deflect the questions, in the apparent hope that the next one would be about some aspect or other of the software architecture.

Let’s examine some of these questions. Is Mr Bell in fact authorised to record other people? What is the usefulness of all this data? How could it be accessed, except ultimately by replaying the subject’s life? (And whilst recording the act of replaying, presumably.) How might carrying recording devices around one’s neck influence one’s behaviour, outlook, attitudes towards others - not to mention theirs towards us? Wouldn’t you be very careful of what you said and how you looked in the company of Mr Bell?

I’m sure you can think of several more questions besides. But here’s one I find particularly intriguing: for this project Bell employs an assistant, Vicki Rozycki. Now if Ms Rozycki wanted to document her own life, would she need her own assistant? And what does this say about class, status and memorability?

Not only the Bill Gates of the roughly hewn The Road Ahead, but also the far more sophisticated and compelling thinking of Nicholas Negroponte’s Being Digital needs to be borne in mind here if we seek to shine a light on the ethos behind Bell’s project. Negroponte’s understanding of technology is unwaveringly utilitarian, and strictly at the service of business and progress, but with a curious whiff of old aristocracy about it. He talks of the digital butler and of electronic cufflinks communicating with each other via low-orbiting satellites, and you can imagine him typing those words in the executive suite of an hotel somewhere in Asia while the permanently over-excited Brit who covers the business world for CNN brays about something or other on the telly. Indeed, while Negroponte proclaims that we are all digital, Bell chimes “I am data”^[3]. But no matter how frustrated one might feel at the shallowness of their thinking, there is a significant extent in which it is also true. The executive of a transnational corporation or the high-powered consultant who advises the rich and powerful and sits through corporate meetings in which minutes never fail to be taken; who regularly consumes in-flight entertainment and receives daily emails in the hundreds – this is the kind of person who has an ‘output,’ whose words are always recorded, who is data.

At the same time, Bell undercuts this narrative by eschewing Negroponte’s constant reminders of his own status and (self) importance, claiming instead:

I’m not particularly interesting. I’m just typical of what you should be able to do.

Isn’t this then also a case of the ‘lowering of the threshold of description’^[4] that Foucault so ably describes? It used to be that only the rich and powerful could aspire to have their lives documented. When Winston Churchill was born, for instance, his (aristocratic, powerful, connected) family duly began to collect and record all manners of information about him in the very reasonable expectation that he would one day be biographised, and the Churchill biography by his son Randolph and Martin Gilbert contains indeed a truly phenomenal amount of documents about the man. But Bell proposes to bring a Churchill biography in every home, one for every member of your household in fact. And you should be excited about this, because... because...

Because why? I honestly couldn’t tell you. Bell and his collaborator Jim Gemmell open a paper on MyLifeBits by claiming that ‘[h]uman memory can be maddeningly elusive. We stumble upon its limitations every day, when we forget a friend's telephone number, the name of a business contact or the title of a favorite book.’ And yes, I too get frustrated when I forget a phone number. But surely the answer is not to record everything - that way lies the paranoiac hypermnesia of Funes - but rather to make a note of the things that seem important, including phone numbers.

I am convinced that Bell’s solution is in fact the answer to a whole different question. And perhaps a clue can be found here, in a document that reveals his involvement in another project at Microsoft Research: Digital Immortality. The abstract, from October 2000, reads in part

Both one-way and two-way immortality require part of a person to be converted to information (Cyberized), and stored in a more durable media. We believe that two-way immortality where one’s experiences are digitally preserved, and which then take on a life of their own will be possible within the this century.

Please wait: the saving of Gordon Bell is in progress.

(To be continued.)



^[1]Bill Gates, The Road Ahead (New York: Viking, 1996), p. 303.
^[2]In Species of Spaces and Other Pieces, translated by John Sturrock. Harmondsworth: Penguin, 1997, pp. 244-250.
^[3]Slide of Gordon Bell's presentation at Victoria University of Wellington, New Zealand, on May 29th, 2006.
^[4]Michel Foucault, Discipline and Punish – The Birth of the Prison (London: Penguin Books, 1997), pp. 77-78.

Gordon Bell, Jim Gray. 'Digital Immortality'. Association for Computing Machinery, 2000.
Gordon Bell, Jim Gemmell. 'A Digital Life.' Scientific American, February 2007.
Vannevar Bush. ‘As We May Think.’ Atlantic Monthly. July 1945, pp. 101-108.
Michel Foucault. Discipline and Punish – The Birth of the Prison. London: Penguin Books, 1997.
Bill Gates. The Road Ahead. New York: Viking, 1996.
Nicholas Negroponte. Being Digital. New York: Knopf, 1995.
Georges Perec. ‘Attempt at an inventory of the Liquid and Solid Foodstuffs Ingurgitated by Me in the Course of the Year Nineteen Hundred and Seventy Four.’ In Species of Spaces and Other Pieces. Harmondsworth: Penguin, 1997, pp. 244-250.
Alec Wilkinson. 'Remember This? A project to record everything we do in life.' The New Yorker, May 27th, 2007.

7 Grams

It's amazing how much easier it is to be taken seriously if your name happens to be prefaced by the words 'Harvard' or 'physicist', let alone both. Thus earlier this month Harvard physicist Alex Wissner-Gross managed to get the British Sunday Times to repeat his claim that a typical Internet search generates about 7 grams of CO2, a little less than half the amount produced when you boil a kettle of water. In its first incarnation, the article was subtly entitled 'Google and you'll damage the planet', later revised, once it transpired that the claim was - how shall I put it? - a little overstated, into the somewhat softer 'Revealed: the environmental impact of Google searches'. It turns out that the average Google search generates only 0.2 grams of CO2, which is one 35th of the original claim, and one 75th of the emissions released by boiling a kettle of water. Google took polite exception, and offered on its official blog the above more reasoned estimate, later accepted by the Times in an update to the original article.

Case closed, then? Storm in a teapot? I would say not quite, but regardless of that I find the story interesting for a number of different reasons. Firstly, there is the initial headline of the article. Think about it: Google and you'll damage the planet. It says so much about the poverty of the public discourse over technology, doesn't it? Go back to the inaugural post of this blog (go on, you know you want to) and you'll find that it was sparked by an almost identical alarmist phrase: 'The Trouble Started with a Google Search'.

It's not just that it's daft to equate Internet searches with one of the search engine companies (Google probably doesn't mind that, in spite of their claims to the contrary). It's equating the act of searching for information, or parsing the results of such a search, with the medium on which the search is conducted that is misleading. It's not so much 'Google and you'll damage the planet', as 'search for information and you'll damage the planet'. Going to the library or consulting in person a public archive instead of using your home computer is likely to produce quite a bit of CO2. And books, ah, making books releases a lot of CO2 and destroys trees to boot. Best avoided altogether.

Besides: why not 'Make tea and you'll damage the planet?' Is it because we are already wise to the evils of kettle-boiling? And I'd like to know a little more, while we're at it. How big is this kettle? Are the hobs gas or electric? How was the energy produced and transported? So on the one hand it's basically the same ridiculous feeling I associate with using horse power as a unit of measurement: nobody is going to convince me that in a race between a Ferrari 360 Modena and a cart pulled by four hundred and one horses, the cart is favoured to win. But in this case we're not even remotely comparing like for like. Pit a search engine versus looking for the same kind of information offline, and I might be more inclined to listen.

But we knew all this before, didn't we? Put Facebook, Google, Beebo into your average media outlet search facility, and it's like fashioning your own generator of random alarmist headlines, or of sweeping and unwarranted leaps of metaphor. La Repubblica claims the top entry today with the headline 'A rischio sul web la memoria globale' (The World's Memory at Risk on the Web), rehashing the always popular - and dismayingly reductionist - identification of the sum of online content with the collective memory of humankind. We're all familiar with this sort of thing.

But perhaps the work of Professor Wissner-Gross, in spite of the out-of-whackness of the initial claim and the enduring conflict of interest (as Jason Kincaid notes, he co-founded this company, which helps websites reduce their carbon footprint) deserves to be looked at again. For taking a Google search as the unit of activity on the Net is not unreasonable, and measuring its environmental impact is not without its practical and conceptual benefits.

Because let's face it, when it gets to the kinds of technology that manipulate symbols - as opposed to boiling water, or transporting your body somewhere - it's easy to lose sight of the material side. Your computer is a solid chunk of stuff including, but not limited to, acrylonitrile-butadiene-styrene, silica, quite a bit of gold and a number of known poisons. It uses energy at work and even at rest - something that the many people who keep their machines on at night seem to be either unaware or dismissive of. But the experience of using it is so out-of-bodily at times that we (I'm making assumptions here) forget about the chunk of stuff altogether, as well as the whizzing electrons and the underground cables and the service personnel and your power bill and the greenhouse emissions and the economy that sustains the whole thing etcetera, etcetera. We get lost in the seduction and the abstraction.

But using a computer makes things happen, not only locally but also remotely. Operating a Google search fires up simultaneously a number of Google servers in different, undisclosed locations. That's what your average 0.2 seconds and 0.2 grams of CO2 buys you, and it's before you've even clicked onto one of the results. Multiply that for your daily activity, then multiply that by the number of computer users on the planet, and it's bound to amount to quite a few KWs (kettle-worths). Just how many? The best current estimate is that the IT industry has the same impact on the environment as the commercial aviation industry. It's not a net impact, of course, we're still faced with the impossible task of subtracting the activities that we'd be carrying out without computers, including no doubt fewer, but individually far costlier, searches for information. But it is something real nonetheless, a thing with weight, as are the cathedrals of discarded computers leeching hazardous substances that dot the planet.

It pays in so many ways to be aware. There are people out there who are seriously entertaining the notion of a digital afterlife, the transfer of mental life onto the machine. Next time one of these witnesses knocks on your door, sit them down at the kitchen table and ask them to account for the cost of maintenance of the electronic heavens. Demand figures and projections. But it's not just about nagging others - as I do weekly in this space - but also gaining a sharper sense of our own entanglements and interdependencies, and of the some time measurable weight, down to the gram, of our habitual actions. Or, as John Ryan and Alan Durning put it in their excellent Stuff: The Secret Lives of Everyday Things, it's about learning to manage your wake.


But for now I'm going to kindly ask you to hold this thought until the next instalment of this most slow-going series on the ecology of memory. I promise to fail to get to the point then too.



Bonus Track
My Life with Legumes
An occasional series*
1. That One Time When I Was Woken Up by Chickpeas

The following really happened to me two weeks or so ago. I was in bed, around midnight, when I was woken up by a click-clickety noise coming from the kitchen. I got up to investigate, but couldn't locate the source of the noise. Nobody seemed to be trying to force the back door. Could it be a mouse or rat inside the wall? Nah: the noise was much too crisp, like when you pop the bubbles on that irresistible packaging plastic.

It took me half an hour to work it out, by which time I had been driven halfway round the bend: the sound came from a bowl of chickpeas that Justine had left to soak in water. Slowly their skins were starting to crack open.



(*) inspired by the fact that so many people find this blog while searching for information about beans.