But I soon ventured beyond the design literature in search of answers. I was sure there was a book hiding away somewhere that had everything I needed. As it turns out, what I needed was scattered across a whole lot of books.

I have spent much of the next two decades trying to wrangle it all into one place, so today I'd like to walk you through a little bit of what I've learned in that process. When graphic designers use the word printer today, we're usually referring to an office machine that spits out a physical copy of something we've composed on a screen. We are aware that printers can also be people, of course, but after the web and digital printing, we tend to spend less time on press checks than we used to.

However, the word printing originally encompassed knowledge and skills that extended far beyond the point of contact between ink and paper. Early printers were often also type founders, publishers, and booksellers. Even as the craft became more specialized printing still involved typesetting and composing pages, and this often extended to a role in writing.

According to union typesetter and print historian Henry Rosemont, newspaper printers in the mid 19th century had a broad but informal education in things like language, history, and geography, which would allow them to write articles on the basis of telegraphs that just had a handful of keywords. So early print workers held a strategic position in the public dissemination of public discourse, and that was simply impossible without them. You could not print text in multiple without hiring a typesetter or a printer.

They often took advantage of this position to educate themselves and to advocate for the interests of their trade. In addition to their obligatory literacy, they had access to the press as an organizing tool, both of which were extremely rare for industrial workers of the era. Journeyman printers became the first workers in the United States to go on strike just after the Revolutionary War, and print workers went on to play prominent roles in revolutionary movements across the world over the next two centuries.

The printed book was, if not the first, then certainly one of the clearest early examples of the world of standardized, mass produced commodities to come. The craft of typography has thus always been linked to the double edged sword of automation. Printing innovations could render work less taxing and dangerous, but the primary motivation for their adoption was usually the reduction of labor costs, which meant layoffs or slashed hours.

Print workers often found themselves, like many workers of that erade, in the paradoxical position of fighting against technologies that promise to ease the burden of their work. Though Johann Gutenberg's 15th century press had scarcely changed in the intervening years, the first decades of the 19th century saw rapid expansion and industrial concentration. Iron construction and steam power fundamentally changed not only the shape of the machines, but the entire work process that fed and maintained them.

Amid all of this industrial acceleration, the process of setting type by hand became an exasperating production bottleneck. Several attempts were made to mechanize the process, with most ending in expensive and spectacular failure. For example, Mark Twain invested and lost a small fortune on a machine called the page compositor, which we're seeing here, which had more than 18,000 distinct parts.

As Twain is said to have boasted shortly before the company went bankrupt. The page could quote, work like six men, and do everything but drink, swear, and go on strike. Very clear that a lot of the kind of wonderful technologies that have made our work possible really came out of a context of suppressing and, and sort of trying to stop strikes or break them.

Then, as many of you know, in 1886, Otmar mergenthaler a german engineer working under contract with some of the largest New York newspaper companies, presented his employers with the first successful attempt, which molded lines of type from liquid metal. The finished lines could then be stacked into columns and locked into page layouts for the press. While the linotype was an expensive and somewhat risky investment, it delivered on its promises of labor cost savings and in time contributed to a dramatic enlargement of the size and circulation of the periodical press.

The International Typographical Union, which I'll be calling the ITU, had been founded in 1852 to represent manual typesetters, and the union was initially slow to appreciate the significance of machine composition. For a few years after the New York Tribune unveiled the linotype in an 1889 issue, Union newspapers often published reassuring reports on the poor quality of its output. A February 1891 article in the ITU newspaper Typographical Journal, for example, argued that the machines were overrated and little more than toys for print capitalists who had no practical background in the trade.

Another article published the next month, however, summarized a report filed by union delegates. After witnessing an improved line of type in action, the delegates and I think it's very interesting how you can pick up their kind of sensitivity to typographical form. Here.

They addressed earlier defects in typographical form and noted that the new machine produced lines perfect in spacing and with a clear face. At the same time, however, they pointed out the difficulties workers encountered in transitioning to a sharply accelerated work process with a complex ensemble of moving parts. The report cautioned against panic but acknowledged that far reaching changes were looming and closed by recommending that the union organize for control over the new process.

In 1892, the New York Tribune and the ITU local here in New York signed a contract that put the newspaper's machine composition under union jurisdiction. During the next decade, mechanized typesetting began to overtake hand composition, primarily in newspaper print printing plants. While there was still a limited need for the manual composition of headlines, advertisements, and other display applications, the new work process soon touched off an employment crisis across the printing trades.

Younger typesetters scrambled to learn machine composition, while thousands of older or more narrowly trained workers fell through the cracks. Those who could secure employment had assessments added to their dues, providing thin relief to the unemployed. There were well founded hopes that a steady expansion of print volume would eventually compensate for the loss of jobs, but it became a foregone conclusion that some number of men would be sacrificed to what one article called, quote the latter day wizard of progress.

The ITU began to push for shorter workdays in order to claim a share in the benefits of these new technologies. As machine composition became the norm in the early decades of the 20th century, the ITU tightened its grip on the linotype, which helped it to grow into one of the most powerful unions in the country. However, the new century also brought a number of unexpected threats to the labor process.

Teletype setting mechanisms were developed to run linotypes directly from punch tape like a player piano. Justifying typewriters enabled rudimentary typesetting while bypassing hot metal, and countless attempts at phototypesetting came and went, though these initially made more of an impact on display typography than running text. In many cases, the development of these new technologies was explicitly spurred on by employers desires to get around what they call the union problem.

In 1947, for example, the Chicago local began a nearly two year strike against six Chicago newspapers, including the Tribune. And during that time, the Tribune ordered enough veritype or justifying keyboards to fill an office, and then simply put its existing clerical staff to work, breaking the strike by setting type. The resulting paper had an uneven appearance, as you can see, and made some embarrassing missteps, including the infamous Dewey defeats Truman headline of November 3, 1948.

So this is something set by strikebreakers while the line of type operators were demonstrating outside. But it did hint at the possibility that a newspaper might someday be produced without linotype machines and without a union. Even probably more importantly, linotype composition remained a norm at us newspapers throughout the mid 1960s, when the ITU began to make concessions on the use of computers and other technologies in exchange for measures that protected existing employees pensions.

It was only in 1964 that the New York City local signed a contract allowing teletype setters into the printing plant for the first time, on the condition, however, that the employers had to pay 100% of the profits resulting from that machine into an automation fund for workers that would eventually be displaced by it. This time, the new machines did not simply alter the printing process. Increasingly after this period, they dissolve typesetting into word processing.

A centuries old gap that had separated writing from printing and dissemination was beginning to close, and this gap had been the very ground on which the ITU stood once more. In this time, the newspaper industry led the way in automation, and again the union scrambled to either train people in the new processes, rearrange working hours, or encourage early retirements. Though the ITU had enjoyed substantial control over print production in cities across the US and, for a time, parts of Canada as well.

That's why it's the International Typographical Union. It would be inaccurate to say that this control was absolute. During the 1970s, especially, printing firms began investing in new photo typography equipment.

Such new machines arrived in a chaotic jumble during these years, and it was not uncommon for print shops to go bankrupt after investing in products that became obsolete almost overnight. Confident in their existing arrangements, the ITU decided not to organize these new shops, which disproportionately hired women at a fraction of the union wage. Here are some examples of the kind of industry advertising at that time that really played on this gender division between the old and new form of typographical work.

At the dawn of the Linotype era, union printers had protested the monster, as they called it, that had taken their place at the type case. Here was a machine that eerily replicated their movements without need for food or human dignity. If we stand in the shoes of these obsolete tradesmen, we can see the modern design software that we use every day from a slightly different angle, as an accumulation of knowledge passed down through the hands and minds of countless typesetters over centuries.

This is a common refrain throughout the history of the ITU and organized print workers more generally. And it was vividly rearticulated by a union typesetter who wrote a song sort of at the end of the Linotype era, just as that era was dying, Carl Schlesinger wrote, computers and cathode tubes replaced us, but by God, they'll not erase us because we taught them everything they know. If you've seen the Linotype documentary, there's a great re recording of this song.

Of course, the technologies by themselves didn't erase these outmoded typographical workers. But for most of the next 40 years, the design discourse did. When I arrived at design school in 1996, I was vaguely aware that the past decade had opened up a new chapter in graphic design history.

In 1983, Philip Meggs had published a still canonical textbook on the rise and fall of modernism in Europe and the United States. Right at that same time, designers in graduate programs and experimental journals had begun to flirt with linguistic and literary theory. In 1984, of course, the Macintosh computer arrived, popularizing experiments with pixelated type until Postscript enabled high resolution output in 1986.

What no one told me was that the same year an organization called the International Typographical Union closed its doors. At that time, it was the longest running union in the United States history. This brief account of about 150 years of typographical history in the US should sound a bit familiar, even if the details have long been obscured in the design literature today in discussions of everything from touchscreen kiosks in the service industry to artificial intelligence in the creative professions, anxious references to robots recall some of the first protests of early print workers.

And so here I'm thinking of things not just like AI, but also the gig apps, like fiverr template apps, like canva. There's a weird sort of, from my perspective, in the particular sort of generation that I came up in, it was at first a huge wave of sort of euphoria about technologies which may come for us eventually. And I think this is sort of an ideal framework to kind of look at those technologies again and really think about why they were invented in the first place.

I think a lot of times, in the nineties and early two thousands, our attitude toward the tools that we used was that they had been invented to sort of save us time or to make things more efficient for us, or to give us new creative tools. Really, this history is very much a history of the suppression of strikes and the disciplining of labor. So I think that's a really important aspect of it to make sure that we sort of keep in mind as we look at what some of the next steps for these technologies might be.

I want to say one thing about the whole sort of AI question, which is that in labor history, it is fairly rare for a technology to sort of immediately obliterate an entire category of work. What automation tends to do is to automate and replace specific kinds of tasks within a job description. And what it tends to do, in a kind of broader sense, is not to completely replace workers again, but to rather cheapen and discipline labor.

And one of the things that I've noticed in reading about that era of, of technological change in the early 20th century is that both the companies that were developing the machines and the employers who were threatening to use them really relied on hype about what these machines were capable of, way ahead of the machines actually catching up with it. And this was actually another way of disciplining the labor force without actually having to invest any money. And I think that's something we should be really careful about when we look at some of the claims that are being made about these new technologies.

Whether they can, whatever they can do decades from now, is a different question. But I think that part of that hype is also about driving down the value of people's work, all, really, all across the economy. So if there is a kind of a general message here, I would say that one thing that this research has taught me is that when I see a new technology or an old technology.

What I try to do is to see labor, to understand the kind of labor that it replaced, to understand the kind of underpaid labor that generated the profits that were invested to make the machine in the first place. And I think that's what we should really think about as we look at these so called AI systems, which are really just scraping thousands of years of human expression that's been uploaded to the Internet and sort of laundered into a form that seems like it's being spontaneously generated by a machine. In closing, I'd just like to say that the way that I've presented this is obviously something that comes out of trying to retrace my own steps, trying to understand the sort of significance of the practice at the time that I stepped into it in the mid to late nineties.

But this work and the bit of sort of exposure I've been able to have. Bye. Circulating some of this work has put me in contact with a lot of people who are coming to very similar questions and conclusions in different national contexts.

So I think that this could be a really promising way of looking at design history in a broader history of work and capitalism. And for those of you who find this at all interesting, I would ask you to join me. Thank you.