reading because internet

I bought Gretchen McCulloch's book after she promoted it on Bluesky

Because Internet: Understanding how language is changing is Gretchen McCulloch’s 2019 book about language use on the Internet. It’s pretty good! I give it eight out of ten. The author deliberately writes in a style that is on the informal side, just a touch too chatty for my liking. She’s some years younger than me, and my stylistic taste perhaps some years older than me, and I twitch at seeing “cool” in an edited publication.¹ But McCulloch enlivens her text by playful toying with words, a skill not unique to linguists but one very common in those who analyse language for a living.

The font for the most part is not worthy of comment, but its exclamation mark has a serif, a little tick at the top to the left. Here is one under a microscope:

Apologies for the imperfect focus; I was holding the hand-held microscope in my hand.

I thought that this was going to be a minor mystery for me to solve, but McCulloch seemingly anticipated my curiosity. On the very last page (recto), there is “A NOTE ABOUT THE TYPE”, telling us that it is Berling LT Std, created by Karl-Erik Forsberg. Personally I think that exclamation marks look better without serifs.

The first chapters

The introductory chapter sets the scene. Linguists are typically most interested in studying everyday speech, which can be difficult and time-consuming even when recorded. It is much easier to study the vast amount of text in newspapers and books, but formal prose does not necessarily reflect the language as she is spoke. Informal writing has traditionally been difficult to study because personal letters and postcards rarely become public.

The Internet changes this state of affairs, with vast quantities of informal writing in chat rooms, forums, and social media, all ready to be analysed for hidden regularities and structure. A delightful example presented is of keysmashes, like as;lkdfjas;lkdf. Dvorak keyboards have vowels on the home keys, and several Dvorak users told McCulloch that they don’t do keysmashes anymore because strings like aosnetuhoasnteuh don’t get recognised for what they are.

The book’s title uses a construction – because followed by a noun – that became popular in informal writing on the Internet in the 2010’s. My sense, not backed up by systematic study and perhaps wrong, is that prepositional because is not as common now as it was a decade ago. Slang can change quickly, and the author is aware that some of what she writes is a snapshot of informal English in 2019.

But the book is not just a time capsule. Chapter 2, forty-six pages long, includes some discussion of geographic variation in language and its study both pre- and post-Internet. I wished for more depth: I was too lazy to go to the endnotes and look up the references. I wanted to see the maps of German or French pronunciation and vocabulary from 100 years ago; more analysis of the popularity on of ikr and ^_^ etc. across different American cities based on geotagged tweets.

(In a later chapter, McCulloch says that “Twitter is overstudied compared to every other social network because Twitter makes it very easy for researchers to collect a large, random assortment of tweets and search through them by date posted.” Those were the days!² Elon raised the cost of that API from free to $42,000/month last year; there is now a $5,000/month option.

At least there is now a decent mass of people using Bluesky and its fundamentally open, public protocol with a firehose API.³)

In a brief comment on language change, McCulloch says that the reader was “probably in early adolescence” when they learned about swearing, which hints at the existence of a cultural difference between Canada and Australia.

As though punished for wanting more depth in chapter 2, I found chapter 3, forty-six pages long, a little tedious with its taxonomy of Internet users (by era, simplifying: Usenet/MSN/Facebook/born into it; there are some sub-categories based on how extensive people’s Internet usage is). The highlight was a discussion of separation characters, especially the ellipsis, which remains popular among some older people. McCulloch finds some old postcards and (informal) recipe cards in which three dots or a dash were used as separators, arguing that this was a space-efficient way to indicate a break on a small piece of card. Today’s chat streams are not space-constrained in this way, and so young people simply press enter, but some of the older generation have taken their old habits into the new environment.

There is also an analysis of lol, one of those entertaining pieces of linguistics that clearly explains what you recognise to be true but had never articulated. I’ve spent my adult life resolutely resisting the use of lol, even though it is solidly part of my mental idiolect: I often think the word. My avoidance while typing is “a conscious choice, fueled by… [a] sense of social identity,” as McCulloch writes in a different context. I am not the sort of person who writes lol. Five years after the book’s publication, I saw a tweet saying that you could tell a message was from an old person if it had lol in it. So it goes! But it is probably too late for me to look young again.

Chapter 4

Chapter 4, forty-six pages long, is on expressive typography, and it is my favourite part of the book: a blend of observations and analysis pitched perfectly to me, a non-expert reader interested in linguistics. McCulloch continues with some more discussion of ellipses and the potential for misreading the tone of an email or chat message: young people often don’t expect to see an ellipsis ending a sentence, and therefore imbue its presence with (perhaps ominous) meaning. More famously, ending a message with a full stop can lead to “Why are you angry with me?” responses from young people.

McCulloch exhorts us to try to be understanding of other people’s practices and standards of what constitutes polite communication. While agreeing with this, I nevertheless feel some quiet resentment at the hostility of younger generations to the trailing-off ellipsis. It was long part of my natural writing style, ably representing a tone that I might want to convey, and I’ve felt the need to train myself out of it because I see people saying that it looks terrible. My resentment is in no small part caused by agreeing with the young people on one level: the trailing-off ellipsis feels good for me to write, but when reading, I feel that a sentence ending with an ellipsis lacks oomph….

Internet style guides from the 1980’s and 90’s had to discourage use of all-capital letters, which were then and now read as shouting, because some early computers only showed capital letters, and some users new to Usenet carried over that typing practice. Repeating a letter at the end of a word for emphasis long pre-dates electronic communication. I’m not sure I learned anything as such from the descriptions of ~***~sparkle punctuation~***~, but it was supremely entertaining to read.

I had previously associated exclamation!compounds with LiveJournal and Tumblr fandoms in the early 2000’s, and had no idea of their origins: bang paths that were sometimes used for email addresses prior to the standardisation of the username@domain.com format. The exclamation!compound was borrowed into Usenet discussion of The X-Files, and subsequently propagated through fandoms.

There is a long discussion on ways to communicate irony in text form, with McCulloch doing a brief survey of historical suggestions for new punctuation marks to denote this, none of which ever caught on. Today informal writers have ironic capitals, hashtags, sparkles, etc. at their disposal.⁴ Some aspects of the non-ironic default style have changed with smartphones, which usually auto-capitalise the first word of a message.

But, crucially, [typographical] irony requires this baseline in the first place. It required us to develop a set of typographical resources for indicating straightforward types of voices, like shouting and enthusiasm, before we could creatively subvert them.

Chapter 4 was terrific, worth the price of the book alone.

The last chapters

The remaining chapters are not forty-six pages long. Chapter 5 is on emoticons, kaomoji, and emoji, with McCulloch arguing that the latter’s rapid adoption across the world is due in part to parallels with gesture in speech. A fun little passage was on the “noseless” and “noseful” variants of the old emoticons :) and :-) etc., the former being tweeted more to younger celebrities and the latter to older celebrities.

“The members of the Unicode Consortium had definitely not signed up to become the smiley faces people,” writes McCulloch, evoking sympathy from me for the poor unicode people.

The facial expressions are by far the most popular, and yet there’s an important way in which they’re not like our ordinary kinds of facial expressions…. [Y]ou can’t involuntarily give off an emoji.

Chapter 6 is on the changing nature of conversations in the chatbox era, and contains some interesting history of chat interfaces before their standardisation, as well as some history of how the telephone changed greetings: hello moving unevenly from a telephone greeting into general usage.

Vestiges of hello’s attention-getting function can still be heard anytime you experience a faulty connection, however; you can say “Hello?” to test the signal, but “Hi?” somehow doesn’t sound right there.

Caller ID and the ubiquity of text messaging have changed norms surrounding answering a phone call, which used to have a sense of urgency.

Chapter 7, the final long chapter, is on memes, and is decently entertaining, but as someone who only consumes memes, I don’t share the author’s evident enthusiasm for making them. And for all the fun memories of lolcats that I have, and for all the worthiness of documenting Internet culture for future historians, I am happy to leave the lolcat bible to a forgotten Internet past.

Of her examples, McCulloch writes that she “picked Philosoraptor and Scumbag Steve above because I’d expect anyone familiar with memes to have heard of them”. I regret to report that I would not have been able to name the Scumbag Steve picture or say what it is supposed to convey. My only knowledge of the meme comes from a reference to it on Twitter during the 2016 Democratic primary, which read as particularly aggressive to those of us unfamiliar with the meme.

McCulloch closes by suggesting that instead of English being symbolised by books, such as a static dictionary, we should think of it as the Internet, a vast place for linguistic innovation and variation. Overall, a fun book for those who sometimes read text online.

Coda

I had intended to study some linguistics when I started undergrad more than two decades ago. But I misread UQ’s rules for electives in the BSc and wrongly concluded that LING courses wouldn’t count to my degree (I would in fact have been able to take 16 units’ worth). So my interest in linguistics has always found its outlet in popular writing, and more recently YouTube.

Reading Because Internet rekindled my old interest in the subject, and I idly looked up what online courses there might be in Australia. I knew as I searched that I wouldn’t actually go back to uni, spending my leisure hours after work for the next six years getting a BA for no reason but personal interest. But maybe it was worth a look? A recent Geoff Lindsey video got just a little bit too technical for me in the arguments about the most appropriate symbols to represent my Australian vowels. Wouldn’t I like to learn phonetics properly?

Well I searched and I learned that a BA in Australia now costs more than $14,000/year full-time, the Morrison government having more than doubled the cost after changes that were announced in pandemic-era mid-2020. Arts now costs as much to study as Law, which offends my sensibilities. That science and engineering now cost substantially less than the humanities feels like an unjust privilege.

I assume that academics screamed, but I haven’t paid attention to higher education policy for years. I wouldn’t have been able to stop the changes, but I wasn’t even listening. Perhaps there will be a partial reversal in 2026.

1. There may be a gender angle here as well:

Research in other centuries, languages, and regions continues to find that women continue to lead linguistic change, in dozens of specific changes in specific cities and regions…. The role that young women play as language disruptors is so clearly established at this point that it’s practically boring to linguists who study this topic.

2. My example from 2014 used the search API rather than the random stream.

3. This sentence comes with both sincerity and awareness that I am trolling because of the recent ~controversial~ creation of a dataset of one million Bluesky posts.

4. I’ve disliked the use of ‘TM’ for this purpose since my Slashdot days.