Research in our time offers a welcome flood tide of investigation into how cognitively modern human beings use their basic mental operations to think and act. With luck, it will not ebb. It could become standard, in the way that calculus, once it arose, abided. This tide offers special emphasis, crucial for this issue, on the cognitive origins and operations of language and literature, and in particular on the ways in which systems of multimodal forms can be deployed to prompt for mental operation.

The preceding article, by Antonina Harbus, is a superb contribution to this research and a shining example of its overarching design. Harbus presents key analyses that illuminate her historical subject matter and that moreover have wide and important scope across language and literature:

1. 1. Human beings, dealing with vast and otherwise unmanageable conceptual networks and relations, need to be able to chunk and compress them to create bundles of meaning tractable within the limits of their basic mental operations.

2. 2. Forms of expression can be designed to prompt for such chunking and compression.

3. 3. These expressive forms are typically multimodal.

Indefinitely many fascinating, pyrotechnic examples of such compression are everywhere:

In this example, analogies and disanalogies across decades of time, unknowably many points of space, complicated long threads of causation, and hundreds of agents are compressed to the everyday human scene of two people interacting: the 12-year-old boy reacts to the adult's decision, and the adult, recognizing and acknowledging the boy's evaluation, decides to disagree. This is an instance of the very common compression template known as fictive interaction (Pagán Cánovas & Turner Reference Pagán Cánovas, Turner, Pascual and Sandler2016): in the compressed blend, the boy and adult are simultaneously jointly attending to the adult's decision; this compression is prompted for by the use of 12-year-old me, current me and the present-tense third-person word is. This compression is unified, tractable, memorable and useful; but no reader is deluded about whether the actual 12-year-old boy, outside the compression, in the vast and diffuse conceptual network, can see the future. This fictive interaction compression might have been further cultivated into fictive conversation, a common subtemplate of fictive interaction, as in, 12-year-old me keeps complaining, but I keep plugging my ears, la-la-la-la.

A range of non-pyrotechnic, everyday constructions that prompt for compression is reviewed in Blending in Language and Communication (Turner Reference Turner, Dąbrowska and Divjak2015). Those constructions include, among others:

1. 1. Words, like safe, danger, lucky, accident, mistake, gap, dent, missing, detour and many others, like decedent. Safe, for example, deployed in reference to some situation, prompts us to activate a mental web corresponding to that situation, blend parts of it with the frame of harm to create a single, compressed scene of harm in the situation, understand that the compressed blend is counterfactual with respect to the original situation, and now blend the original situation and the counterfactual blend so that the counterfactual relation between them is compressed to absence of harm as emergent structure in the new blend. Absence of harm is an element that is not in the original conception of the situation and not in the frame for harm. It is an emergent structure that comes from compression, and it is the focal concept for the construction of meaning. This compression involves conceiving of absence of harm as a property in the blend – a compression that involves the outer-space counterfactual relationship between mental spaces. Negatives like no can perform similarly: At our Sunday sale, get no sales tax.

2. 2. Morphemes, such as over and under in words like overfishing and undernourish.

3. 3. Change predicates, as in dinosaurs turned into birds and his girlfriend gets younger every year (this example from Gilles Fauconnier, p.c.), in which analogies and disanalogies across very many mental spaces are compressed to change for an identity. The analogies are compressed to the identity and the disanalogies are compressed to change for that identity.

4. 4. Blended syntax, as in risk in, risk on, risk to and phrases like load the truck with hay and smear the wall with mud (Fillmore & Atkins Reference Fillmore, Atkins, Lehrer and Kittay1992).

5. 5. Nominal compounds (land yacht), French double-verb causatives (Pierre fait manger Paul).

6. 6. A great range of XYZ and related constructions, e.g. These flame retardants are the asbestos of our time. (See more discussion below.)

7. 7. Viewpoint blends, e.g. past tense + proximal now, e.g. She now saw that he was lying.

Rhetorical traditions also picked out patterns of form that prompt for compression, e.g. hypallage and transferred epithets, such as loud man, sleepless night, happy morning, bitter afternoon, early grave, guilty pleasure, grateful memories and Thomas Gray's ‘The ploughman homeward plods his weary way.’ In the compression, e.g. the man has the property of being loud even though the relationship outside the compression is one of cause and effect: the man sometimes performs actions that result in noises that others judge to be loud. In the blend, he has the property of being loud even when he is completely silent.

At the heart of Harbus's analysis is a generic integration template that results in one of the most useful patterns of compression in human conceptualization: the cycle. Consider our conception of a day. In our experience, there is actually just one day and then another day and then another day and then another day, in a sequence that stretches out indefinitely, forward and backward. The days in that sequence are all quite different. They do not repeat. If we woke up today and it was exactly the same as yesterday because it was in fact the same day in every detail, we would be sure we had lost our minds. And then it would not even be the same day, because yesterday we did not think we had lost our minds. Day after day after day indefinitely, with all those differences between days, is too much to comprehend, too much to fit inside working memory, too much to carry around and manage. It is not mentally portable. So we blend these different days into a conception of a cyclic day. We do so by using a general template for creating cyclic blends. There are analogies and disanalogies across the different days in our experience. The analogies are packed into one thing in the blend: the day. The disanalogies are packed into change for that thing: the day is cyclic; it starts over every dawn and repeats. It is important to recognize that the cyclic day blend is not just an abstraction. There is new stuff in the blend that is not in any of the individual days in the mental web to which the blend can be applied. Indeed, almost no blend consists exclusively of structure that is equally shared by all the mental spaces upon which the compressed blend draws.

The same generic compression pattern works to structure a poem as the repetition of the poetic line. Just as we can refer to the day and not mean a specific day, so we can refer to the Greek epic poetic line and not mean a specific line. Of course, just as for days, it is not the case that all the many details of each day are projected to the blend: the poem does not consist of the identical line's being repeated. And, just as for days, there is emergent structure for the poetic line: it repeats, which is structure for the compressed poetic line but not for each poetic line in the decompressed network. And just as there are elements in the compression that vary across individual days (there is a slot for my morning coffee in my cyclic blend, but that does not mean that I have coffee every morning), so there is variation available in the compressed, blended poetic line, but highly structured and limited variation. In Classical Greek epic, the poetic line is a dactylic hexameter, five dactyls followed by a spondee, but a long syllable can substitute for two short syllables (but only rarely in the fifth foot), and the last syllable of a line can be an ancepts, i.e. a short syllable counted as a long. The poetic line contains an opportunity for a pause (caesura), and the sound pause often aligns with a meaning ‘pause’. The caesura is always between two words, in the middle of a foot, either between the thesis and the arsis or within the arsis itself, between the two shorts. And so on. Latin epic has a very similar compressed poetic line with highly structured and limited variation. For each, we have the compressed notion of the ‘poetic line’, which repeats. This creates a very compressed structure for the epic, no matter its length, which is evidently exceptionally useful for both the production and the reception of oral epics.

And so it is with the Old English poetic line, as Harbus analyses. The central feature of this ‘repeating line’ is the alliteration (phonemic more than phonetic) across the caesura, with, as she discusses, the possibility of double alliteration. We will return to formal alliteration. Digression 1: other poetic traditions in other languages use the same generic compression template but with a different kind of repetition across the formal caesura – e.g. a kind of acoustic heraldic symmetry about a midline, so familiar in art and architecture. The human body, with its gross right–left symmetry, is our constant lived embodiment of repetition across a midline. I do not read Egyptian, but adduce Wilson (Reference Wilson and Frankfort1949: 50), who cites the Egyptian ‘strong sense of balance, symmetry, and geometry’ that ‘comes out in [Egyptian] literature, where the best products show a deliberate and sonorous parallelism of members, which achieve dignity and cadence’. He illustrates with a text whose speaker is an Egyptian king:

• Give heed to my utterances/ hearken unto them.

• I speak to you/ I make you aware

• That I am the son of Re/ who issued from his body

• (Wilson Reference Wilson and Frankfort1949: 50)

Sweetser (2005: 34) says of poetic lines as compressed blends:

Digresson 2: Compression of succeeding elements to a ‘repeating’ ‘cycle’ can operate at any level. Beowulf, for example, is often described as a battle cycle combined with a funeral cycle: the battle passages treat Grendel, Grendel's Mother and the Firedrake (dragon); the funeral passages treat Scyld Scefing, Hildeburg's brother and husband, and Beowulf himself. Some add the Lay of the Last Survivor to the funeral cycle. There can be internal structure to the ‘repeating’ unit: for example, the funeral consists of the preparation of the body, the wake, the funeral proper and the funeral feast.

Alliteration additionally, as Harbus discusses, prompts the hearer to perform a compression of relations. The alliterative relation of similarity of sound can prompt for compression to tighter vital relations, as discussed in Fauconnier & Turner (Reference Fauconnier and Turner2002). The result of this invitation to compress can be the construction of a tight relation of cause–effect, part–whole, identity, group (e.g. kin), time connection, space connection and so on. It is not a weakness of alliteration that it does not dictate the resulting compression. This is common for forms that prompt for compression. The ubiquitous XYZ construction (Necessity is the mother of invention) and its many related constructions (Fauconnier & Turner Reference Fauconnier and Turner2002; Turner Reference Turner, Dąbrowska and Divjak2015) leave almost everything about the resulting compression unspecified. Language, words, forms do not mean, any more than chess pieces on a chessboard mean. What is meaningful are the ideas that a human being constructs in response to forms. All forms themselves are prompts for human beings to use mental operations they already possess to work on things they already know. Rhyme has a similar operation: why is it so embarrassing when children on the playground say to two peers, ‘Bob and Ma-ry, sitting in a tree / K-I-S-S-I-N-G’? Why does ‘Marky, Marky, Marky, he's such a snarky’ bite so hard for children and even adults? Because the similarity of sound invites a compression to a non-acoustic inescapable meaning relation. For a full discussion of rhyme and its logic, see Sweetser (Reference Sweetser2005).

The most important point to make is that the form–meaning pattern of alliteration, although it has special and brilliant nuances in literature, is anything but restricted to literature. At the 2016 Democratic National Convention in Philadelphia, former Secretary of State Madeline Albright, endorsing one of her successor Secretaries of State, Hillary Clinton, for the Democratic nomination for President of the United States, said:

The alliteration prompts us to make an integrated meaning packet of whatever is alliterated, so that ‘smart power’ comes to consist of all the coordinated meanings we construct in response to those four words beginning in /d/. It also prompts us to think of other words beginning in /d/ to add to this list, and to wonder or dispute about them: dynamism? duty? diligence? decency?

Traditions of drawing distinctions between language and literature – stretching from Aristotle's Poetics to Northrop Frye's Anatomy of Criticism and beyond – have been invaluable and informative. But language and literature form a continuum. Thought, language and literature are not silos but part of the seamless operation of the cognitively modern mind.