| A Fan Of Freedom |
(Read the book here).
Sam Williams's biography of Richard M. Stallman, Free As In Freedom, is one of the most clear-eyed and sensitive biographies I have ever read. I have known RMS since we were teenage SF fans back in the mid 1970s. I've met his family. I've collaborated with RMS and argued with him and hung out with him for more than twenty-five years. For all that events since 1998 have forced us into a sort of rivalry, RMS and I remain among each others' oldest friends. Still, this book helped me understand him better.
That needs to be said at the start, because I'm going to spend a lot of the rest of this essay criticizing FAIF. This criticism isn't intended to be hostile. Sam Williams's version of RMS's story is, in my opinion, not wrong so much as it is missing some crucial perspective, some of which I hope to supply in this essay. I'm giving it this effort precisely because I think the book is of high quality, and worth the best amplification and critique that I can render both as a historian and as a friend of RMS.
And yes, the book helped me understand my own history better, too. There are parallels between my life and Richard's that I didn't understand the full significance of until Williams's narration pushed me to think about how unusual those shared experiences are. So this essay is going to be a little about me, too. Sam has been threatening to do my biography next...
When Sam sent me the FAIF manuscript in draft, a first reading quickly revealed that it had a problem I had expected, one that has dogged all accounts of Richard's career since Steven Levy's Hackers in the 1980s. The author has been in part seduced by his subject — he has bought into too much of the mythology RMS created around himself.
I suppose it's difficult to avoid this trap with RMS. He is an intensely charismatic man who has refashioned his entire life around a Big Message. He has a tendency to project the drive and certainties of his later self, the fully developed crusader/evangelist, back on to his earlier years. RMS's account of his own development tends to be streamlined, narratized. Not false, but tending to pass lightly over the false starts and uncertainties and growing pains and hesitations that were visible to those of us that knew him way back when.
It's diagnostic that Richard can no longer recall with certainty the year or the other person involved in the CMU printer-driver incident that set him on the path to "free software". Once the story had been absorbed into Richard's personal narrative of prophecy, struggle, and mission, the actual time and circumstances of the story became unimportant to him and forgettable.
The "real" story of RMS, or at least the story I witnessed, is a bit less epic and more human. I can remember a Richard who hadn't figured out what he was about yet, a fairly ordinary-looking young man with short(!) hair who between 1977 and 1980 or so chased several of my girlfriends and (despite RMS's later report of his consistent lack of success with women) caught at least two of them. I don't see that Richard in Sam's book. Instead, going on RMS's report, FAIF has given us something that is largely the RMS-approved version of his earlier self.
Perhaps it could not have been otherwise; I sometimes wonder if even RMS knows the difference any more. All of us narratize our lives; we all reinvent our own histories to some extent. But one of the perils of reinventing yourself as a prophet/visionary is that you can end up forgetting who you were before the vision. It's a possibility that has worried me about my own life since fame was more or less thrust on me in 1998.
Richard's reinvention of himself began twenty years before this was an issue for me. It was already well under way when, a year before the GNU Manifesto was issued in 1985, Steven Levy described RMS as "the last true hacker". That's an unhelpful myth, about which I've been annoyed with Levy ever since. At minimum, it should read "the MIT hacker community's last standing survivor".
This is more than a historical trivium. I was a hacker at the time, and I knew many others. We were not at all affected by the "Symbolics War" that so scarred Richard because the ground we were standing on was different — we were Unix guys. We wouldn't get seriously hurt by commercialization until a few years later after the AT&T divestiture. The MIT group was important, but they were not the only hackers in the world. Levy's hagiography of Stallman implicitly consigned the rest of us to the memory hole.
Richard's own tendency to emphasize the printer-driver incident over the consequences of the Symbolics War tends to downplay the cultural context in which he operated — one consisting not merely of other MIT hackers such as Steele, Gosper, Abelson, Moon, et. al. but of hackers in other traditions — the Unix guys, the Xerox PARC crowd, the SAIL group.
Sam took Levy's "last true hacker" description out of the FAIF draft when I explained why it was bogus. But the narrative bias that goes with it, especially the tendency to portray Stallman as the sole link between what came before him and the post-1985 hacker culture, is still present. The real untold story here isn't Richard's individual development of a sense of mission — that prophet-out-of-the-desert tale has been rehearsed way too many times. It's the way in which RMS's choices grew organically from a hacker culture that already existed, and which informed RMS's values and the range of his choices on many levels.
I can remember that pre-Stallman era. Most of the rest of us old-timers have given up trying to remind the world that the hacker culture did not in fact collapse in 1980-1983 only to spring from RMS's brow reborn with the GNU manifesto; I sometimes feel like a last holdout myself on this one :-). But RMS's choice to model the GNU operating system on Unix reflects on a technical level the cultural fact of a pre-existing community.
In FAIF, Sam quotes Rich Morin: "The community reaction [to the GNU manifesto] was pretty much uniform. People said, 'Oh, that's a great idea. Show us your code. Show us it can be done.'" He's right; that's how we all felt. I remember thinking that GNU seemed like a great idea, but that this shaggy dude I knew from SF conventions seemed an unlikely candidate to carry it off. He wasn't one of us. He wasn't a Unix guy.
Richard drew much more from that culture than he seems to remember nowadays. I myself was the person who first suggested to him, at a Boskone in 1983, that Emacs ought to be the GNU project's first product. We'd been friends for nearly five years by that time. It is utterly characteristic of RMS that he says he doesn't remember that conversation, but believes my report that it happened. As with the printer-driver incident, once the idea had been assimilated into his personal sense of mission, the circumstances under which it formed became irrelevant.
And it's not just RMS's cultural context among early hackers that FAIF tends to miss, but his formative experiences among SF fans as well. Again, one of the most interesting things about RMS's story is the respects in which it is not unique, but representative of themes which recur constantly in the lives of people like us. For "people like us" read "gifted-to-genius geekboys who acculturated within SF fandom and closely related subcultures in the 1970s and 1980s".
Seth Breidbart, who Sam quotes and who I've known as long as I've known Richard, is another member of this cohort. So is Don Hopkins, who RMS credits with inventing the term "copyleft". Either could tell you that the constant punning, the little mindfucks like the "Impeach God" button, the painstaking ethical reasoning, the particular kinds of emotional uncertainty RMS is prey to, and many other signature traits Sam ascribes to RMS as an individual were actually considered pretty normal and unsurprising where we came from.
Even the hand-kissing. Richard probably picked up that gesture from people in the Society for Creative Anachronism, same as I did. I still do it sometimes when I'm at SF conventions. It's within the envelope of normal behavior there.
And beyond hand-kissing — Richard's "love but not monogamy" credo is not the individual invention Sam portrays it as either; it reflects the values and practices of the first community in which Richard found a social niche. The practice of polyamory has been common among SF fans since I became involved with fandom in the mid-1970s, though it would not be named or develop an explicit ideology until the mid-1990s. Whether Richard himself remembers that he owes this aspect of his values at least partly to his history in SF fandom or has allowed himself to forget it is an interesting question — and one that parallels his relationship to the hacker culture at large.
We were all of us misfits out to impeach God. Rebels against mundanity, stupidity, the ordinary. We wore short hair when the fashion was long and long when the fashion was short. Like RMS, we were — and are still — essentially compelled to go where intellect and imagination and ethical conviction leads us, regardless of whether that's comfortable or socially acceptable.
What distinguishes RMS (and myself) from Seth Beidbart and Don Hopkins and Alan Walker and Mark Miller and other members of this pack is not that our character is essentially different, nor even that our ability is necessarily greater. It's just that, for whatever reasons of historical accident, RMS and I went public with it. We have projected our interests and personalities outside the subcultural venues in which they are considered normal.
All of us more or less always expected people to find our intellects impressive, but I'm often bemused at the extent to which our other shared traits strike people as exceptional. What other way is there for us to be? What else could we be but driven, rigidly ethical, ruthlessly analytical, anti-authoritarian, idealistic, careless of normal social rewards, countersuggestible, etc., etc. It's in our wiring. We're self-analytical, too; we can see that wiring. We have little choice in what we are.
Maybe we all have Asperger's Syndrome. Like RMS, we tend to consider the prospect dispassionately and somewhat dismissively (his tendency to be unsparing and clinical about his own problems, when he recognizes them at all, is also typical of our kind). Labels are just labels, and the boundary between normality and aberration is one we are accustomed to thinking of as fluid and often a matter of definition.
In fact, from RMS's point of view it's the rest of the world that is broken. Bereft of either his intellect or his moral vision, ‘normal’ people stumble through routine lives, never waking up to the larger issues that frame their existence, and yet somehow drawing from each other animal comforts that are mostly denied him. All of us felt that way in the darkest passages of our adolescences; it went with the territory, as surely as knowing what a glider gun was, and being able to sing Tom Lehrer lyrics from memory, and reading Scientific American and thinking the Mensa test was a pretty good joke. Richard, more than any of us, stayed there into his adult life.
The parallelisms go beyond just psychology or attitudes, though. It was pretty normal for us to fling ourselves prodigy-like at mathematics or science, find we lacked either the discipline and maturity or some other quality needed to make it there at the level of our aspirations, and fall back on programming instead. Like Richard, I aspired to be a mathematician — gave a research paper at an AMS conference before I graduated high school, took grad-level courses as a college freshman — but burned out. Others in our cohort could doubtless tell similar stories. But like Richard, we have all tended, then and now, to pass over failure lightly in telling our histories. We, even more than most people, because we were afflicted by the sense that we should not have failed.
Like Richard, I too stopped working on proprietary software after 1985. But I was quiet about it where he was noisy. He had found his banner, his ideology, his place to stand; it would be another decade before I found mine. In the meantime, I argued with his. I felt the same pain he did; I knew there was something monstrously wrong about a software-production system that alienated me from the product of my creativity, buried my work under secrecy and bad management decisions. But I thought Richard's anti-propertarian dream was crazy, a nightmare that would lead to starving programmers and no good result. We had a number of late-night arguments at conventions over this.
The others in our cohort who had become programmers adapted better than Richard and I did, settling into conventional jobs. One of us coped by taking lots of hallucinogens, a conspicuous exception among a group that (like Richard and myself) had generally stayed strictly clear of drugs even when they were socially acceptable in the 1970s. Another one became a guru of such inaccessible fearsomeness that his bosses at the New York financial house where he worked dared not mess with him.
One way or another, the rest of us created environments around ourselves that enabled us to avoid or evade the central problem RMS was grappling with — how to fix the proprietary-software system that was crushing the creative life out of us? We couldn't sign on to Richard's eccentric messianism. We knew him too well. We were too much like him. But I think we all felt the same pain he did at around the same time in our lives.
Again, my theme here is that Richard's experience was nowhere near as exceptional as FAIF makes it sound. Many of the brightest hackers all over the world were coming, in their individual ways, to the same cusp point. Richard's genius was that he exteriorized his personal journey as a plan for a movement, then mythologized it and took it on the road — providing a model for me when I felt it necessary to do the same thing ten years later. It could have been any of us, though. Could have been Seth, or Alan, or Don, or Oz, or Mark, or Karl, or possibly any of a half-dozen other geekboy geniuses I knew in SF fandom whose names have escaped me.
FAIF quotes Stallman as follows: "I was there, and no one else as far as I could tell was going to do it," recalls Stallman. "So I had to do it. I was elected." While it took individual vision and moral courage to act out this feeling as thoroughly as Richard did, the sense of obligation, of being elected by the fact of one's own capability wouldn't have surprised any hacker of that day or today. Hackers often react like this. See something that needs doing, do it — don't count the cost. In this respect as in many others, RMS was as much a product of the hacker culture as a shaper of it.
I don't say these things to take anything away from Richard. just to place him in the context that FAIF is missing. He did act. He did change the world, if not perhaps in the exact direction and degree he intended. RMS put his stamp on the hacker culture so pervasively that it is sometimes difficult to tell where Richard's reinvention of himself ends and the cultural matrix around him begins. I admired him greatly for that even as I disagreed with him, having no clue that within a few years I would be doing the same sort of thing myself.
And that points up perhaps the biggest difference between us, beyond theory or ideology. Richard would never have described himself the way Linus Torvalds and I later did, as accidental revolutionaries. Linus and I had to be dragged into that role and we still aren't entirely comfortable with it, but for the man who became RMS, it was purposeful revolution from the beginning. Because when you're a genius in the secularized-Jewish tradition that produced Spinoza and Marx, that's what you do — you seize on a big moral idea and redeem the world with it.
In Terry Pratchett's Discworld fantasies, one of the continuing themes is the peculiar power of myth. Stories seize people and nearly compel them to act out a predestined drama; courage, on the Discworld, often consists in breaking out of the script. Pratchett, of course, was reflecting in a fantastic way the human urge to make narrative sense out of life, for human action to have meaning and plots to have resolutions. Pratchett's novels are immensely popular among hackers.
FAIF buys the whole script for RMS's drama, the eremite hero/prophet whose individual moral vision and self-martyrdom changes the world. That story-line requires that, despite intense suffering and rejection, the prophet's vision ultimately triumph — the greater the suffering, the more complete the eventual victory. So FAIF requires the canonical ending — that RMS's "star has grown", that his moral crusade is stronger today than it has ever been.
The facts are otherwise. The failure of the GNU project at its central goal (production of the HURD kernel) became undeniable in 1995-1996, and as a result RMS has been steadily losing influence to others since then. In 1995 nobody could even name anyone but Stallman with a credible claim to lead the hackers. Now, all but the hardest-core FSF supporters would rank Linus Torvalds above him in influence, a good many would add Larry Wall to that list, and not a few would add me.
As an index of the change, do a Google hit count on the terms "open source" and "free software" (with "GNU or GPL or FSF or Stallman") as qualifiers; this is a phrase notoriously prone to false positives. Then count pages that reference both. Now think about the results. I performed this test twice, in late 2001 and September 2002 (later figures would not be comparable, as Google changed the implementation of its search rules in early 2003):
| Term | 2001 | 2002 |
| open source | 2,860,000 | 3,020,000 |
| free software | 1,330,000 | 1,890,000 |
| both | 2,010,000 | 2,660,000 |
| free software only | 850,000 | 360,000 |
| % of total | 38% | 16% |
In other words, the number of webpages that endorse RMS's "free software" label and his refusal to countenance "open source" dropped from 38% to 16% of the whole set during that year, and not just in market share but in absolute numbers as well. This was a crude test, but it fits what I observe on my travels. It's not a picture of burgeoning success, or even of a long march to victory. The community has spoken. RMS's propaganda tactics have been found wanting and largely abandoned for more effective ones.
Furthermore, RMS obviously agrees with me that the "free software" message seems in the process of being completely eclipsed by the "open source" message to which he objects so vociferously. He complains about this constantly in public interviews, and in recent years has a tendency to get emotional and upset about it whenever we're in the same room.
Once again FAIF seems to be ignoring the facts of the situation in favor of a storyline that makes RMS the lone mythic redeemer destined to triumph against a Philistine world. Reality is more complex than that; indeed, it is more complex than that 38%-to-16% drop indicated by crude content analysis might suggest. Because there is indeed a sense in which RMS is winning, but not at the exact game he wanted to play.
RMS's artifacts — GCC, Emacs, the GNU General Public License — really have changed the world. The process of open, collaborative development he did so much to help invent is triumphing. His code and his license have succeeded; it is only his rhetoric and moralizing that have failed. The tragedy is that RMS himself values his moralizing more than his code.
Had it not been so, had RMS been a bit less consumed by his big moral idea and a bit better salesman, the hacker culture would not have landed on me the job of saying what he could not. I might have been a fully paid-up member of his crusade, rather than winding up as an unwilling rival.
I often think I would have preferred that outcome. Perhaps then RMS's story would have had the happy, redemptive outcome FAIF projects. But that's not the history we got. Instead, I think RMS will never be satisfied with the victory he gets, even as it transforms the world.