“. . . In that Empire, the Art of Cartography attained such Perfection that the map of a single Province occupied the entirety of a City, and the map of the Empire, the entirety of a Province. In time, those Unconscionable Maps no longer satisfied, and the Cartographers Guilds struck a Map of the Empire whose size was that of the Empire, and which coincided point for point with it. The following Generations, who were not so fond of the Study of Cartography as their Forebears had been, saw that that vast map was Useless ….” [1]
—Suarez Miranda, Viajes de varones prudentes, Libro IV, Cap. XLV, Lerida, 1658
We cannot possibly communicate with perfect accuracy and precision [2]. The information contained even within the simplest idea is boundless when all scales and contexts are considered. Dealing with potentially huge amounts of information tests one’s ability to package and communicate it to receivers, who are themselves challenged by their abilities to interpret. Effective communication therefore necessarily sacrifices precision and sometimes accuracy as well. Just like Suarez Miranda’s map, useful pronouncements are compromises.
In or out
In the 1950s and 60s a nation-wide program in New Math was established in North American grade schools [3]. This introduced students to classification and assortment – the very basics of set theory. New Math better prepared students for higher education and careers in science, mathematics and engineering. But more importantly, the foundations of New Math turn out to be central in processing information in every-day life [4].
Yes-no, in-out, grouping can enormously simplify complex problems and in so doing ease communication. For example, it would be complicated (and pretty boring) to classify the reasons for why a sample of 100 people are in a given place (let’s say a shopping mall) at a given time (3pm). We could create sets such as ‘shopping’, ‘accompanying a shopper’, or ‘other’. All of the people in the mall would fit into one or more of these. Complex becomes complicated should the classification include additional layers, such as ‘how long do you expect to be at the mall?’. The concatenation of such potentially endless information is useless and confusing (except of course if you are studying people in shopping malls!). Worse, receivers may understand something different than what the emitter intended. And even worse, different receivers themselves differ in their understanding of a given message. More on this later.
For example
I attended a protest of the US actions in Grenada following political unrest in 1983 [5]. The evening of October 25th in Berkeley to be exact [6]. Never having attended such an event, I was eager to see what it was like, and was joined by a colleague. I was struck by the huge contrasts among the marchers. A few were shouting slogans, and by their frequent fleeting inquisitive stares at one another, apparently competing to gain followers who would repeat their chants. During one of the many lulls in the cortege, my colleague (who was several years older than me and said he had taken part in many similar marches) turned to me, and said “watch this!”. He ran up to the front of the group and started shouting his own slogan, with a much louder, booming voice than the other leaders. For a time, he prevailed and led several hundred people, but eventually decided he had succeeded in his little experiment and blended back into the group. I left the demonstration at the steps of Berkeley City Hall, as several slogan shouters tried to reinvigorate the crowd.
My take on that Berkeley evening is that the 'group' was a complex mix of attitudes, unequivocally centered on protesting the general idea of military intervention. (No one could have possibly known all the details of the Grenada intervention, since that information would have been filtered with national security in mind). Weaving a precise [7] description of the Berkeley demonstration would be more of interest to sociologists than to the lay public, the former who rather want to understand reasons for the military operation. Journalists covering the event would have interpreted it as a march, a protest or a demonstration, and where along the inevitable precision-communication trade-off to package their story. Marchers relating their experience afterwards would be subject to the same trade-off, but being parti pris, their takes would likely contrast (as did mine) with those of bystanders and journalists.
The Trade-Off
The simplest grouping is speaking of two things as if they are one. We have no problem seeing apples and oranges are not one, but when compared to a car, the former are joinable into fruit. And if we look close enough, apparently identical things do have differences. An even should two objects be absolutely identical down to the atom, they cannot possibly occupy the exact same space at the exact same time.
In other words, grouping necessarily jettisons variation, and in so doing follows a trade-off between precision and communication. Why? On the one hand, simplifying through grouping facilitates both emission and reception. On the other, grouping rounds-off and generalizes – it therefore loses information [8]. But, even should we want to be more accurate and precise in our pronouncements or search for what lay beneath those of others, we risk losing the audience in the former and make unfounded assumptions in deciphering the latter. These forces push communication to err on the side of simplicity [9].
Norms
The proclivity to group is a type of norm. We do it and expect others do it too. Yet grouping norms are not particularly well-honed insofar as we are not identical in how we treat information: each person has her own comfort zone on the precision/communication trade-off surface. For example, organized protests were so unfamiliar to me that during the Berkeley march I dialed my information reception to a minimum. I didn’t focus at all on the content of the slogans, but rather on the physical (fist-pumping) and vocal (screaming and shouting) behaviors of other marchers.
Variation
Norms work because they are no-nonsense, expected-to-be-accepted behaviors. They cut through time-consuming detail and obviate debates and confrontation. But normative behavior – and grouping norms in particular – does not address variation in understanding the content behind the norm. This variation may stem from observational and processing errors, or differences in context between individual receivers. For example, the events on January 6th 2021 at the US Capitol Building were variously called a coup d’état, a siege, a riot, and a protest, and the participants anything from domestic terrorists to patriots [10]. A good fictional example of observational variation comes from the 2008 political thriller Vantage Point, where an assassination attempt is observed and seen differently from different points in an arena [11].
Emitters and observers have different experiences, proclivities and abilities to create and to make sense of information packages. What this means is not only the emitter packages a message under the constraints of the tradeoff mentioned above, but also needs to account for the makeup of the audience. To further complicate matters, the audience members are each potentially exposed to multiple emitters. And if this were not enough, variations may be generated at each juncture in a chain of communication [12].
Given what is ostensibly optimization strategies in the honing of precision, accuracy, and context in packaging, could this have its roots in Darwin?
A Darwinian Interpretation
Constraints, tradeoffs and optimization are indicative of an evolutionary process [13]. Is Darwin at the root of how we package and interpret information? To the extent that reproductive success – that is, Darwinian fitness – is influenced by how we as a highly social species communicate, selection on and evolution in traits influencing communication success are to be expected. Otherwise said: one’s attention to accuracy and detail is sculpted by costs and benefits [14]. An across-the-board optimum is however not expected, since communication traits are complex [15] and variation will manifest both from person to person in a given environment, and from environment to environment for a given person. As we know only too well, this variation can be huge.
Limited capacity
Brains are limited in their abilities to gather, process, and communicate information [16]. Do I say what I think? Does the receiver understand a message the same way I intended it? What does the receiver actually understand from what they hear? The forces molding communication have analogy with those behind social interactions themselves as revealed by Dunbar’s Number: the number of meaningful social contacts that an average individual can maintain [17]. Dunbar’s Number is molded by the benefits of social interactions, subject to the costs of potential conflicts and of dilution every time a new contact is added [18]. All else being equal, having more contacts is better, but as their number increases their quality eventually decreases due to less time spent with each interactant, resulting in lower reproductive success and therefore selection for some intermediate number of meaningful contacts. Similar to Dunbar’s Number, the optimum communicated package is arbitrated by trade-offs involving cognitive capacities.
Environmental mismatch
Our brains are adapted to the social environments experienced over countless generations. The associated gene- and culture-based behaviors are not fixed however, and rather give us the flexibility for individual and social learning, which, in turn, enable dealing with informational challenges [19]. Yet as much as humans show great capacity to cope with complex problems, the past few decades have tested limits. The environments we now encounter, being multi-scale and multi-dimensional, push what was challenging in our recent-past to the sometimes unfathomable in present-day [20]. In particular, the huge increase in both meaningful and meaningless interactions in conjunction with all types of media has meant that we are routinely exposed to far more information than we can possibly deal with, and even more than the most powerful computers can ‘intelligently’ process [21].
One of the byproducts of the mismatch between adaptive information processing and present-day environments is ambivalence to what in our evolutionary past (and even in our lifetime past for the older amongst us!) would have been Darwin-appropriate reactions. What would have elicited a strong reaction a few decades ago is now met by ambivalence. Such temporal shifts actually might have an evolutionary basis, if it were the case that we are under selection to allocate costly effort only where needed. We learn with time which packages are routine and need little attention, and those that are out of the ordinary and require tuning-in.
Selectivity
There’s an evolutionary explanation for why we focus attention on and react more actively to extremes: survival. Survival means obtaining sometimes scarce resources and dealing with occasional life-threatening situations. As survival factors become commonplace we learn routines to seamlessly deal with them. Focusing too much on the mastered and mundane saps time and energy from the challenging and important. The evolutionary legacy of this is our selective attention to the amplified and the uncommon (loud sounds and bright colors, art, music and sport that move us, but also exposure to singular events, prestigious individuals or fancy products), and to information relating to pressing concerns.
Such attention bias has important implications for packaging information. To what extent should I summarize, contextualize or alter information to gain a busy listener’s attention? Do I first lure the listener in, observe her reaction, and only then decide how much detail to add? The challenge of effectively transmitting messages goes some way to explain how we package information, but how does this percolate through a population of emitters and receivers to influence expectations, that is, norms?
Normal keeps amplifying
In the world of information different inputs compete for attention, storage and transmission [22]. In the world of competition, persistence and growth requires doing something better or different than others. Major media networks – such as CNN, MSNBC and Fox News – are the epitome of how information and competition interact to hold and grow audiences. Tune-in and look at the attention such networks give to every detail. These and other for-profit information services depend on commercial sponsorship, which, in turn, relies on viewer attention. Viewer growth means keeping a step ahead, and analogous to adaptive mutations and recombination in biology, this is fostered by tactical tweaks and strategic leaps. Growth means more R&D, more change, potentially more growth, and so on. Through time, broadcasting physiognomy evolves, and this – that is, the packaging – influences how we-the-public see the world and package information ourselves [23].
Media packaging is driven by survival, power, and often, plain greed. A venerable reference is the 1976 movie Network, where decision-makers on a TV station discover how the extreme behavior of their prime-time newscaster attracts growing numbers of viewers, and with it, commercial, media and cultural success. Again, think of major media networks where sponsors pay on the order of $100,000 per half-minute of advert time [24]. Sponsors need watchers, networks want sponsors, and watchers seek the new, important and exciting. But as we saw above, watchers acclimate. If seen over and over again, what is exciting eventually becomes mundane. This keeps media on their toes to innovate. But to always keep a step ahead, such shifts in media culture need to be incremental. Counterintuitively, they work because they take time to be noticed at all [25],[26].
Miranda’s Maps
This essay was about our proclivity to group diverse information into discrete packages. I then briefly visited the inner-workings of packaging, specifically the unavoidable trade-offs and the possible evolutionary basis for norms and fuzzy optimality.
The main message is that due to constraints we simply cannot escape some degree of grouping. The people who marched at Berkeley, did they all have exactly the same motives, or even vaguely similar ones? What if all did, except for one person? Were they all 'protesters'? I’m pushing this a bit far, but take a moment and think about any simple characterization and the inevitable underlying subjectivity that must exist within that grouping. It doesn’t take much to see how this applies to most everything [27]: single group characterizations of nationality, political or religious affiliation, or more ephemeral groupings such as “supporter of …”, but also advertisement tactics, such as referring to anything new as an “innovation” [28].
So, where does this leave us?
Given rapidly changing social networks and social media, we simply don’t know what package landscapes will resemble in the future. Nevertheless, greater attention to how we interpret packages and package ourselves has the potential to foster more meaningful interactions. The will require a greater appreciation for what we see and who we broadcast to and interact with. But in the meantime... we will never quite understand why a smile is so beautiful.
.........................
[1] From On Rigor in Science – Jorge Luis Borges, 1946. Here is an antecedent from 1895 in Lewis Carroll’s Sylvie and Bruno Concluded "What a useful thing a pocket-map is!" I remarked. "That's another thing we've learned from your Nation," said Mein Herr, "map-making. But we've carried it much further than you. What do you consider the largest map that would be really useful?" "About six inches to the mile." "Only six inches!" exclaimed Mein Herr. "We very soon got to six yards to the mile. Then we tried a hundred yards to the mile. And then came the grandest idea of all! We actually made a map of the country, on the scale of a mile to the mile!" "Have you used it much?" I enquired. "It has never been spread out, yet," said Mein Herr: "the farmers objected: they said it would cover the whole country, and shut out the sunlight! So, we now use the country itself, as its own map, and I assure you it does nearly as well." [2] Whereas accuracy refers to being factual, precision is exactness. [3] Feinstein 1973. 10.1111/j.1949-8594.1973.tb09087.x [4] Grouping is packaging with specific criteria. A related concept is “coarse graining”, which is a level of resolution that produces pattern that can be understood, ignoring underlying detail (https://en.wikipedia.org/wiki/Coarse-grained_modeling). [5] I use the neutral term ‘actions’, although this was widely referred to as an “invasion”. https://en.wikipedia.org/wiki/United_States_invasion_of_Grenada [6] November 11, 1983 The Militant pg.3 [7] Recognizing that increased precision on a given aspect is useful only up to a point, many aspects (i.e., accuracy) effectively multiply the less useful information contained in the full event. [8] Though clever packaging can actually make a complicated object more understandable, and so although some information is lost, what information remains is simplified/less noisy. [9] Effective communication does not always err on the side of simplicity – but, if you give me tons of details, then I’ll likely stop you and ask to get to the point. [10] https://en.wikipedia.org/wiki/2021_storming_of_the_United_States_Capitol [11] https://www.imdb.com/title/tt0443274/ [12] https://en.wikipedia.org/wiki/Chinese_whispers [13] I use optimization to mean the search for accurate and precise communication, given constraints. [14] See recent article by Pennycook et al, testing causes of inaccuracy and possible remedies https://www.nature.com/articles/s41586-021-03344-2 [15] https://en.wikipedia.org/wiki/Reaction_norm [16] See for example Semedo et al 2019. https://www.cell.com/neuron/fulltext/S0896-6273(19)30053-4 on how neural functions are selective and yet flexible [17] 150 [18] This of course has become disrupted with increasing contacts, social media, etc... [19] See Rendel 2011. DOI: 10.1016/j.tics.2010.12.002 [20] Labeling is exemplified by Sorites paradox, where different grades of a (subjectively) labelled state are quantitative measures themselves. For example, the determination of baldness in humans has no objective threshold as measured by the number of hairs on the scalp; however, a male with fewer hairs per unit scalp area could be said to be balder than someone with more hairs. The Sorites paradox highlights why some qualitatively new phenotypic traits are deemed ‘innovations’, and it highlights the role of consensus and subjective judgement in this labeling. It raises the unsolved question of how or whether relative levels or magnitudes of innovation (marginal to transformational) can be scientifically distinguished from one another. See for example, Fisher P. 2000 Sorites paradox and vague geographies. Fuzzy Sets Syst. 113, 7–18. (doi:10.1016/S0165-0114(99)00009-3) [21] Tradeoff surfaces across the tree of life and at different scales (within organism, in populations, etc.) are yet to be elucidated. One might expect that AI will change the landscape of the tradeoff itself and increasingly bring computers to zones where packaging/grouping minimizes content loss, by being tailored to the listener. [22] You will know the “meme” from social media, though it was originally popularized by Richard Dawkins, and has many less-known antecedents – noteworthy being work on the spread of rumors and the landmark 1967 article by Klaus Dietz https://www.epi.msu.edu/janthony/requests/articles/Dietz_Epidemics%20and%20Rumours.pdf [23] https://en.wikipedia.org/wiki/Influence_of_mass_media [24] https://fitsmallbusiness.com/tv-advertising/ [25] This is because too large a sudden change will either be ignored, stymie growth, or result in loss of share. E.g., https://en.wikipedia.org/wiki/New_Coke [26] I modeled this effect, inspired by a nearly neutral mutation accumulation process: https://www.researchgate.net/publication/8213245_A_theory_of_modern_cultural_shifts_and_meltdowns [27] Notable exceptions being pure logic, mathematics, and laws of physics. [28] Stay tuned for a future blog on single number characterizations of complex systems.