 |
| cover art by Raja Nandepu |
Republic of Memory, by Mahmud El-Sayed, previously reviewed on NOAF, is a deliciously rich text, full of chewy, intertwined meditations on memory, legacy, family, society, duty, revolution, collaboration, and identity. Since you all raced out to buy it and read it back when Stew's review came out in May, the details may be a bit fuzzy, so I'll remind you of the key bits here. (If you didn't race out and buy it and read it DO THAT NOW, jeez, how many times do we have to tell you a book is fantastic?)
The story is set on the generation ship Safina, two hundred years and halfway through its journey to its intended colony planet Hurriya, governed by an administration set up during a ship-board revolution shortly after launch. But the democratic, anti-fascist, anti-AI (w00t!) Compact that was put in place is breaking down. Time and and complacency have led to corruption, and disaffected crew have been dipping their toes back in the waters of revolt. In fact, the creeping collapse of the Administration resembles the Safina herself: ageing, suffering worsening technological malfunctions that range from transient inconveniences to deadly shipwide catastrophes. Complicating matters further are the cryogenically sleeping passengers, selected at ship's launch by the now-defunct Network Empire that funded the expedition. These sleepers are intended to be revived at Hurriya to from the first generation of colonists, and although the new Administration has broken from the Network Empire in most ways, this element of their mission is preserved. Socially, the passengers are revered as Ancestors by the crew. But their preservation is eating up resources on a ship that is feeling the ever-increasing strain; and not everyone reveres them. They are, remember, perpetuators of the old order of the Network Empire that everyone now reviles, a symbol of everything both right and wrong about the ship's origin and purpose. Every time I try to pull on a single thematic thread to examine in isolation, the rest of the book comes with it; no one idea can be analysed independently of the others.
Nevertheless, I'm going to try. And because I'm a linguist, I want to discuss how brilliantly language — and, as we shall see, anti-language — is used to support all the other thematic components of this text, from culture to change to rebellion to rebirth.
1. Language and culture on the Safina
We can see this centrality of language from the start, in mind of Iskander Ezz, a Translator by profession. The first thing we learn about Iskander is that he feels like a different person depending on which language he speaks. This idea — that language affects thought — has been studied scientifically by Earth linguists from a number of angles, of which I will discuss two. The first angle, perhaps the more familiar to readers of SF, derives from the Sapir-Whorf hypothesis. This idea, also known as linguistic relativity, was invoked brilliantly in Ted Chiang's novella Story of your Life and later dramatized brilliantly in Denis Villeneuve' magnificent film adaptation Arrival. (It was also invoked less brilliantly in Heinlein's Stranger in a Strange Land.) This hypothesis proposes that the different structural features of the languages people speak direct their attention to different elements of existence, and so modulate their perception of the world.
In its strongest form, the Sapir-Whorf hypothesis often seems to invoke the concept of time, specifically. Benjamin Lee Whorf himself famously claimed that because the Hopi language lacks any capacity to express the concept of time, the Hopi people are incapable of understanding the concept of time as Europeans do. (Whorf 1941/1956). (He was wrong on both counts.) More recently, Chen (2013) proposed that, if a language doesn’t explicitly encode for future tense, the grammatical unity between present and future might prevent speakers from discounting future gains or disregarding future harms. The future is now, Chen argues, and so the potential gains and harms are more immediate. As a result, speakers of tenseless languages they smoke less, watch their weight, use condoms, and invest their money more responsibly.1 In the limit, for the Chiangs and Heinleins among us, it's a short step from this sort of claim to add the SF-sparkle, and write stories wherein learning a language like Martian or Heptapod B, with sufficiently temporally transcendent grammar, grants its speakers the ability to see the future.
Although no one has demonstrated tense-based clairvoyance in the laboratory, there is some reasonably persuasive evidence that language can affect perception in limited ways. For example, the range of basic colour terms in a speaker's language does seem to shape colour perception in small but measurable ways (Thierry et al 2009). If a language encodes agency, its speakers might better remember intentional actors than unintentional actors (Fausey & Boroditsky 2008). Although most linguists agree that strong Whorfianism is not terribly well supported, weak Whorfianism may not be complete bunkum.
An alternative approach to this idea — i.e., that different ways of thinking depend on different languages — relates not to the linguistic structures of the languages themselves, but instead to the distinction between thinking in a native language and a second (or third, or fourth) language. If you have ever learned another language, think of how you might relapse into your first language in moments of high stress. Native languages tend to be more tied to emotional processing than second (or third or fourth) languages, and this difference can affect cognitive processing. Keysar et al. (2012), for example, asked people to choose between two options in both their first and second languages — e.g., selecting one of two treatment options for a deadly disease. Mathematically, the treatment options were equivalent in their outcomes, but they were framed differently across experimental conditions. In one condition a treatment option was presented such that it had a certain percentage probability of saving lives, while in the other the same option was described in terms of its failure rate. Even though the two options in each condition were numerically equivalent, participants were more likely to select the option framed as saving 33% of lives, rather than the option framed as letting 67% of people die — but, crucially, only when they were considering the options in their first language. When they did the same task in their second language, they instead recognized that the outcomes were equivalent, regardless of framing. These experiments were replicated across a variety of languages and scenarios, and the outcome was reliably similar: people were less susceptible to emotional framing in their second language. The authors propose that the mechanism relates to cognitive load: when you're thinking about a language you know less well, you don't have the brainspace to spare for emotional reasoning.
Iskander's split personality could be understood from either of these two perspectives. Arabek, his native language, is rich with different expressions of emotion — especially cursing (oh, especially cursing!). Why shouldn't using a language which encourages emotional expression result in a personality that is is quick to smile and laugh, effusive and expressive? By contrast, in Inglez, the language of ship administration, Iskander's second language, he is formal and proper, as would be expected in a language whose sole purpose is directed towards formal pursuits.
Alternatively, you could imagine that the cognitive load of operating in a foreign language like Inglez would tax Iskander enough that he must tamp down on cognitive indulgences like emotion. (However, since he falls in love with a woman to whom he communicates in Inglez, this seems less persuasive).
But I think the personality-language match here is doing a third thing. Forget linguistic research. Instead, consider Iskander's personalities through a narrative lens. The Safina houses its thousands of inhabitants in different berths, and although its original crew members came from different nations and ethnic groups, the Compact’s First Crew after the revolution decided to group people not by ethnicity or religion, but by language. By grouping people based on voluntarily mutable characteristics — anyone can learn a new language — the crew can guarantee equality and unity, with translators to ease linguistic barriers to cooperative group action. Thus, the different berths speak Arabek, Manda, Canto, Malay, Russ, Nihon, Torkdil2 and others. And to communicate across berths, all are united by Inglez, the language of official administration.
Inglez is no one's native language, but anyone who wishes to work in Administration, or indeed climb above the lowest ranks in other divisions, must learn it.3 But Iskander's role as a Translator also serves a second purpose: in addition to simply conveying meaning, he also serves as a kind of lawyer. He ensures that all the paperwork — for criminal defenses, marriages, births, deaths — is prepared and filed properly. His use of Inglez, then, is fundamentally administrative. No wonder he develops a stodgy, no-fun, stuffed shirt persona in Inglez: not necessarily because Inglez's grammatical structure encourages stodginess, and not because he is less emotional in a second language; but because that particular second language is tied primarily to the culture of paperwork.
And what goes for Inglez goes for all the other language on the Safina. The linguistic variety might have been intended to serve as a mutable grouping characteristic, but the different berths have preserved — or developed, or strengthened — their differences since launch day in more dimensions than just language. Religion, fashion, culture, and values vary enormously across berths, defying the original goals of preserving social mobility.
The fact that the ship's original crew contained so many speakers of so many languages, and yet not English, feels like a pleasing thumb in the eye of English linguistic imperialism. Whatever this future Network Empire was that had the power to dominate the world and build generation colony ships, it did not require English to do it. However, the fact that English — and thus Inglez — was consciously adopted to become the language of status and privilege on the Safina is harder to interpret. Is it a straightforward world-building consequence of what happens when one language is made the language of power? An authorial commentary on the inescapable dominance of English in our current world? A thematic invocation of the tensions inherent in any society where English-speakers specifically control all decisions, and vernacular speakers must learn English to have any power?
The last option is certainly consistent with the unfolding events in the book. By 200 years into the journey, organized crime groups — or covens — are everywhere, spreading violence in their internecine disputes; and even law-abiding crew have started to develop linguistic bigotries, or 'berthisms' betraying the philosophy of social mobility that underpinned the berth-grouping structure. Administrators throw complications in front of crew who want to marry across berths, linguistic and religious fundamentalists are developing with berths. The Compact is breaking down (like the ship), or perhaps was imperfect from the start (like all societies). Chaos threatens; agitators whisper of revolution, and the crew, divided as they are across berths and languages, are losing touch with the shared purpose that is supposed to unify them in their mission. Iskander is different people in different languages. He's not sure which of them is the real him. It's not clear which version of the Safina will survive to see Hurriya.
2. Language and change on the Safina
Languages change over time. As we can see in the spellings of the Safina's languages, Arabek is not the same as its ancestor Arabic; Russ is not Russian; Torkdil is not Turkish; Inglez is not English. This shift is made all the more evident when one of the sleeping Ancestors, Heba Hilal, is woken from hibernation and finds herself confronted with 200 years of linguistic drift. From her perspective, she talks like a hard-boiled detective — appropriate enough, given her previous employment back on Earth as a military investigator. But when we see her from the Safina's crew's eyes, her speech is archaic. 'Grant you mercy,' she says. 'Wouldst thou like to come in?'
And yet not all speech changes at the same rate. Hilal observes that Inglez has changed much less from English than Arabek has changed from Arabic. This is linguistically believable: standardized languages that are formally taught (like Inglez) will remain far more stable across the centuries than casual, vernacular languages (like Arabek). In fact, Arabek itself is diglossic, with different versions used in different contexts. Formal, scholarly, religious Arabek is much the same as Hilal remembers. Only the vernacular Arabek requires her to rely on her still-intact AI-implant to help her with automatic translations.
This split in Arabek on the Safina echoes the status of Arabic in modern-day earth: Modern Standard Arabic is the language of news, politics, and education. It is no one's native language but everyone's second language (does that sound familiar?). Meanwhile, the vernacular Arabic dialects — Mahgrebi, Levantine, Gulf, Egyption, and others — have diverged from their common ancestor to the point that speakers of some varieties verge on mutually unintelligible. Speakers of Gulf dialects, spoken around the Arabian Gulf (e.g., Saudi Arabia, Kuwait, Qatar) can understand each other easily, but have difficulty understanding Mahgrebi dialects, spoken in Algeria and Morocco.4 On the Safina, the Modern Standard Arabic from Earth, like English/Inglez, have retained much of their 200-year-old character. Only the vernacular Arabic varieties have continued to grow and change until they are nearly unintelligible to Hilal's ear.
The shifting rates of language development mirror the Safina’s social tensions between conservatism and change. This parallelism is beautifully apparent in the role of the Ancestors. Is it respect for culture and heritage that leads the crew to maintain the Ancestors in cryogenic stasis, celebrating them on the anniversary of the launch, visiting them and honoring them on ship holidays? Or is it mindless subservience to the last remnants of the brutal Network Empire, whose voracious violence eventually destroyed Earth itself? Is there anything to be gained by preserving the Ancestors? They were selected, after all, for their expertise in geology, meteorology, agriculture, and all the other skills that will eventually be necessary for colonizing a new world. But they also are the last representatives of that same Network Empire, with the technological augments still in place, ideologically devoted to a regime which the ship's crew rejected centuries before. And it should not be ignored that they consume enormous quantities of Safina’s resources, at exactly the time when the ship is itself strained and breaking down. How can the ancestors' potential future contributions be balanced against the equally undeniable dangers they represent — both now, asleep, and in the future, awake? When they wake on Hurriya, they will speak old languages that none of the crew understand. In their mouths, a world that is 400 years dead and gone will return.
3. Anti-language and anti-society
The fracture lines between embracing change and conserving the past fuel both crime and revolutionary fervor among the crew, who coalesce into multiple activist groups, also called covens. These covens both follow the language-based divisions and also cross-cut them. Along linguistic lines, for example, we have the Karls, or Kommunists among the Russ-speaking births, and the more violently-minded Vory; the Haraka in the Arabek-speaking births; the Gokudo in the Nihon-speaking berths; the Zendistas among the Portuguese-descendents. Although these covens have their own primary linguistic affiliations, they must also communicate with each other, both for purposes of illicit trade, but also for coordinating larger criminal or political actions. Crucially, though, they do not communicate in the existing lingua franca of Inglez. They use a different mode of communication: Nupol.
Nupol is described as an argot, a polyglot blend of words from every language on the ship — and some that didn't make it into space — slotted into an Inglez grammatical frame. Russ contributes numbers and a lot of body parts; German or Yiddish's descendant seems to have provided negatives and some obscenities; Portuguese or Spanish or some other Romance language provided some abstract ideas about work and thought and freedom, along with some personal and relative pronouns, and temporal adverbs. Still, each word can be more or less directly with an intended English counterpart. The reason it's possible to learn Nupol is because it is not an entirely novel language. The syntax of the sentences remains familiar. Even the morphology is unremarkable: To cog — to think or feel — inflects like any other verb: Today I cog, right now I am cogging, yesterday I cogged, tomorrow I will cog. Gulliver — head — pluralizes into gullivers. Nupol is Inglez relexicalized, but not regrammaticalized.
Nupol is almost a textbook example of what Halliday (1976) called an 'anti-language'. Anti-languages are the communicative practices of anti-societies. In Halliday's words, "An anti-society is a society that is set up within another society as a alternative to it. It is a mode of resistance, resistance which may take the form passive symbiosis or of active hostility and even destruction" (pg 570). The covens on the Safina are all of these things. Some — the Zendistas, the Gokudo — seem to be content simply to to exist in their own carved out domains, existing in mostly passive symbiosis with the ship’s administration. Other covens take more active forms of hostility, but differ in the degree to which they advocate for active destruction. The Vory want to overthrow the administration and replace it with their own head, but otherwise preserve the social order. The Kommunistas want to go further, and set up a new social order, build a new Compact.
These revolutionary goals make literal another feature of anti-languages. As Halliday argues, every language reinforces the agreed-upon social order that structures its speakers' consensus reality. Every time someone speaks, they must decide what to say, and what to leave unsaid; and every one of those decisions conveys something about the interlocutors' shared reality: This thing is new, and require explicit remark, and so we say it. All of those other things are normal, background parts of daily life, and not worth saying. So we say, 'Hey, your fly is undone' to identify the bit of the world that departs from the norm, without bothering to remark on all the shared knowledge about the convention of going about clothed rather than naked, the standard modes and mechanisms of standard garments, or the particular tabooness of crotch-associated wardrobe malfunctions. In this way, language helps to construct and maintain our shared reality.
By speaking Nupol, the members of the covens, the coves, are not simply communicating across berths that speak different languages, or signalling in-group membership to other coves, or avoiding being overheard by out-group members — although they are doing that too. But more than that, they are also building the new social reality they want to spread: one without the Administration in charge; one run by the covens, structured by coven social hierarchies, governed by coven values.
Because this new social reality is defined in opposition to the existing standard, it is not, in itself, fixed. There are many ways to oppose the dominant order, and as the covens shift their goals and membership and internal power relations, so too does Nupol shift. The Nupol spoken by Badreddine, the leader of the Haraka, sounds old-fashioned to the younger members, and indeed one of them reflects that Badreddine himself is the oldest person speaking Nupol that she has ever seen. To her, Nupol is the language of youth, of change, of agitation.
An early scene in the book illustrates Nupol's quicksilver changeability in action. Iskander’s sister Damietta is aligned with Badreddine and the Haraka, and in this scene she is approached by a former friend called Britva to convey a message to Badreddine. This is the first time they've met in years. In her youth, Britva had been arrested for smuggling and spent time in prison; and following her release she joined the Russ-speaking Kommunistas. Although both women know Arabek, they converse in Nupol, because the substance of their interaction is aligned with coven activity: Britva's message is a proposed alliance between her coven, the Kommunistas, and Damietta's, the Haraka. During this conversation, Britva uses words Damietta doesn’t know, and so needs to explain their meaning to Damietta: strax — itself derived from a Russ(ian) root — means to fear something. In other words, as Britva’s Kommunist-aligned influence spreads through these proposed alliances, she also influences the Nupol of the people she contacts. New words enter the anti-language and spread, reflecting changes in the structures of the anti-society that uses it.
4. The Inglez substrate
I mentioned earlier that Nupol is constructed by relexicalizing Inglez: new words are introduced to replace standard vocabulary, without altering the grammatical frame. But why must the grammatical frame, the substrate language, be Inglez specifically? As I see it, there are two reasons: one trivial, the other metaphorical and delicious.
Trivially, the coves needs to be able to understand each other. The crew don't necessarily know Arabek and Russ and Manda and Torkdil and Canto. (Even Translators like Iskander don't know all the languages themselves. Iskander is actively working on learning Nihon.) What the crew do all know is Inglez, to one degree or another. So if Nupol is built upon Inglez, everyone has access to a recognizable scaffold from which the unfamiliar bits can be puzzled out. (The reader must do quite a lot of this. It helps to know a certain amount of Russian.)
Metaphorically, recall that Halliday describes anti-languages as the language of an anti-society, and describes anti-societies as organizations defined in opposition to the dominant social order. A completely new language would be a thing in itself, able to exist independently of any other social order beyond the one belonging to its speakers. An anti-language, on the other hand, must be built upon the substrate that it exists in opposition to. You need to start with Inglez before you can build a language defined in opposition to Inglez.
The delicious consequence of this metaphor extends further. Inglez is the language of privilege and power, but by providing the substrate of Nupol, it is also the means of unifying the coves who might otherwise struggle to unite their forces to greater action. If the brewing revolution is to be successful, then the new consensus reality, constructed and reinforced by the use of Nupol, will sprout from seeds planted in the earth of Inglez. And what could be more appropriate? On the Safina, the dead are composted to create food for the new generations that replace them. The old consensus reality is dying. The new struggles to be born. Now is the time of anti-language.
——
1 This particular proposal got a lot of attention at the time, although I don’t believe it was ever accepted as more than a cool idea that lacked proper scientific support. Just last year my department — and I'm sure other linguistics departments in universities around the UK — was approached by a banking firm, because they were considering developing a marketing campaign about savings and investments based on this idea. They wanted to get some real linguists to approve it as valid science. I do not believe campaign ever launched.
2 The languages that Arabic, Mandarin, Cantonese, Malay, Russian, Japanese, Turkish, and English become after hundreds of years.
3 Given that one component of the growing corruption in the ship relates to the way plum Admin posts are passed down within families; and given that people who fall in love across berths often end up speaking to each other in Inglez — as Iskander and his girlfriend Taki do — I have my doubts about how long Inglez can remain 'no one's native language'. It's not explicitly discussed, but I suspect a lot of those hereditary admin families and cross-berth couplings probably do end up speak Inglez at the breakfast table with their children.
4 This information comes from my PhD student, who is currently studying the use of intonation to interpret syntactic ambiguities in Kuwaiti Arabic, and the extent to which Kuwait Arabic-English bilinguals can use the same perceptual strategies to interpret prosodic cues in the two language.
References
Berreby, D. 2012. Obese? Smoker? No Retirement Savings? Perhaps It's Because of the Language You Speak. The Big Think. http://bigthink.com/Mind-Matters/obese-smoker-no-retirement-savings-perhaps-its-because-of-the-language-you-speak
Chiang, T. Story of Your Life, in Stories of Your Life, and others. [Tor, 2002].
El Sayed, Mahmud. The Republic of Memory. [Gollancz, 2026].
Thierry, G, Athanasopoulos, P., Wiggett, A., Dering, B., and J. Kuipers. 2009. Unconscious effects of language-speciï¬c terminology on preattentive color perception. PNAS 106:4567–4570. https://doi.org/10.1073/pnas.0811155106
Fausey, C. M., and L. Boroditsky. 2008. English and Spanish Speakers Remember Causal Agents Differently. In Proceedings of the 30th annual meeting of the Cognitive Science Society. Washington, D.C. https://escholarship.org/content/qt4425600t/qt4425600t.pdf
Halliday, M. A. K. 1976. Anti-languages. American Anthropologist, 78(3): 570-584. https://www.jstor.org/stable/674418
Heinlein, R. Stranger in a Strange Land. [G. P. Putnam’s Sons, 1961].
Keysar, B., Sayuri, L. H., and S. G. An. (2012). The Foreign-Language Effect: Thinking in a Foreign Tongue Reduces Decision Biases Psychological Science 23(6): 661 – 668. https://www.jstor.org/stable/41489753
Liberman, M. 2012. Cultural diffusion and the Whorfian hypothesis. Language Log. https://languagelog.ldc.upenn.edu/nll/?p=3764
Pullum, G. K. 2012. Keith Chen, Whorfian economist. Language Log. https://languagelog.ldc.upenn.edu/nll/?p=3756
Sedivy, J. 2012. Thought Experiments on language and thought. Language Log. https://languagelog.ldc.upenn.edu/nll/?p=3797
Whorf, B. L. (1940/1956). An American Indian Model of the Universe. In Language, Thought, and Reality: Selected writings of Benjamin Lee Whorf, ed. John B. Caroll. Cambridge, USA: MIT Press. Pg. 57.
CLARA COHEN lives in Scotland in a creaky old building with pipes for gas lighting still lurking under her floorboards. She is an experimental linguist by profession, and calligrapher and Islamic geometric artist by vocation. During figure skating season she does blather on a bit about figure skating. She is on Mastodon at wandering.shop/@ergative or on Bluesky at https://bsky.app/profile/ergative-abs.bsky.social
