Understanding the use of language in science. Some limitations on abstractions

The common language used in non-scientific environments tends to be largely vague, at times inconsistent, elusive and recursive. What one person is saying or writing can hold almost uncountable different interpretations. Rather than being a defect, such ambiguity and imprecision underpin the success of language as a communication instrument. These features indeed enrich communication, foster imagination and lubricate social interaction. Language is a fundamental component of human life, making us what we are as animals.

Science has a fundamental and undeniable dependence on language. However, scientific language is not perse neutral, objective and free of the aforementioned shortcomings. On the contrary, the language that scientists decide to use is always prone to reflect particular points of view and world visions, and can eventually be biased on ideological, religious or philosophical grounds. The particular language used in a scientific text is a choice that, intended or not, reflects personal beliefs, experiences and points of view. Furthermore, as it is based largely on abstractions that take distance from phenomenological facts, scientific language is prone to become ambiguous, especially if the arbitrary boundaries used to delimitate such abstractions and metaphors are not explicit.

Common scientific abstractions like cells, trees or ecosystems, are used to deal with particular complex phenomena present in reality. However, in the real universe, in a strict sense, there is no such a thing as “a tree”, with a such broad level of generality. What is out there and can be touched and perceived is always a particular tree, very specifically located in space and time that requires to be identified by a myriad of phenomenological characteristics that describe a unique event in the universe: the big, old and thick tree (Nothofagus Pumilio) located in the backyard of my aunt in Southern Patagonia, that I loved to climb. That tree is specific because the above expression is specific enough to refer to one single phenomenon that can be directly related to personal experience and not misunderstood by something else. I climbed “that tree”, not another one. Unfortunately, it is hardly possible to produce scientific knowledge with such levels of specificity. On the contrary, one of the fundamental characteristics of scientific endeavours, i.e., reproducibility, requires the use of abstractions; concepts that cluster large sets of phenomena and are used to propose general principles. Therefore, scientists can not refer to the unique experience mentioned above, rather they use the concept of the “tree” to identify a large group of phenomena that share a set of defined properties and that can be classified as belonging to the group of trees in this case.

Scientific language largely relies on abstractions. Scientists must choose the most appropriate conceptual option to refer to a phenomenon that is always highly complex and difficult to link to particular boundaries. There is indeed a permanent tension between a proper level of abstraction that can foster complex thinking versus the necessary precision that can pinpoint specific phenomena. This is a routinary trade-off that all scientists must face in the process of developing scientific knowledge and communicating it to peers and beyond.

Scientific language must fulfil a set of characteristics to be considered as such. Scientific descriptions require having not only the capacity to be reproduced and tested, meaning the provision of empirical proofs but also, and most fundamentally, the principle of generalization. A phenomenon that can’t be generalized, that applies only to one singular and unique event in the universe and nothing else, is most likely an artistic fact, a beautiful painting or a sculpture, rather than a scientific fact.

Concepts are always arbitrary

The very act of defining an abstraction contains, always, some level of arbitrariness that is context-specific. The tree concept used above requires to be operationally defined to become a proper scientific abstraction. Several phenomenological characteristics like size, canopy, colour, type of leaves, roots, diameter, geometry, etc. have to be used to classify, to differentiate from the overall vegetation types that we call trees. This is always a tricky exercise, especially when such phenomenological characteristics are not unique to trees; indeed, there are other species that at times might seem like trees and might have some of those features while not others, making the tree-identification difficult in reality. My point here is not taxonomic, I’m not interested in highlighting the difficulties of the taxonomic classification of species. My point is the arbitrary assumptions that need to be taken to classify reality, a situation that appears every time we use an abstraction. The reason why this occurs is simple but astonishing: while the complexity of reality is infinite (yes, you have read well, I said infinite) our abstractions are limited by our perceptual apparatus, that condemns humans to perceive only a small fraction of reality and to perceive it in a very specific way. Certainly, the smelly world perceived by dogs is dramatically different from the colourful one that apes like us perceive. Furthermore, the fact that we can hear sound rather than see it with our eyes is merely a result of our evolutionary drift. This very old human limitation was beautifully expressed 2.500 years ago already by Protagoras: “Of all things the measure is Human, of the things that are, that they are, and of the things that are not, that they are not”.

Language evolves and changes

There is also the problem of an evolving language. The scientific concepts we use are prone to change, as language evolves and some expressions that meant something in the past become meaningless or can completely change their original meaning. Such changes in the use and meaning of abstractions play a fundamental role in scientific revolutions. For instance, the aforementioned use of “revolutions” in the sense of “dramatic change” is the prominent actual meaning, which has its origins in the title of Copernicus’s book “On the Revolutions of the Heavenly Spheres”. In that book, Copernicus used the word “revolutions” to refer to the circular trajectory that planets describe around the sun. Today, this circular meaning of the concept of “revolution” is secondary and present mostly in relation to machines and engines. Another good example is Newton’s scientific notion of space and time, which is absolute, meaning that every phenomenon in the universe shares the same space and time and therefore dramatically differs from that of Einstein, which is relative, meaning that every event in the universe possesses its own space and time. However, the layman’s daily use of such concepts remains certainly Newtonian and people have problems understanding that time is a concept that applies only locally, where we are, and not to the whole universe. It is still paradoxical for everyone to think that if you travel at the speed of light and come back you will remain young while most of the people you know will be gone.

There is only one scientific discipline that is not prone to temporal conceptual change: mathematics. Indeed, the concept of zero is the same for us, for Einstein, Newton and Leonardo da Vinci. Mathematics does not have to deal with the temporal changes in its conceptual apparatus, because mathematics itself is “timeless”. However, while relying on conceptual abstractions that are linguistically constructed, all other scientific disciplines must deal with such linguistic changes and therefore revise their use of concepts and definitions.

What to do? Conceptual clarity is the way forward

The first attitude to have appropriate use of scientific language is to always pursue conceptual clarity. Using direct short expressions instead of pretentious long rhetorical paragraphs certainly helps. The routinary exercise of wording-selection to carefully assess the accepted meaning of a word choice, especially when it comes to core concepts of a research field, it is ought to be for scientific writing. A word choice that has been carefully selected will always stand out in the clarity of ideas. As language evolves, it is always good to look at the etymology, the origin and development of words, and understand in which context some expressions were first proposed and used. This applies not only to single words but to full expressions. A good example is Adam’s Smith expression: “the invisible hand”, which is a double reference to the virtue of the market’s self-adjustment, on the surface, but at the bottom is an indirect reference to the invisible hand of God, coming from a highly religious person, and therefore approaching dogma rather than scientific fact.

Some expressions that I always avoid

Doing scientific research in the urban realm (urban ecology) exposes part of my findings to the social sciences where certain terms and concepts seem to be inherently valid. However, several of these largely used concepts and expressions lack the necessary neutrality that scientific language should hold. They correspond to abstractions that are heavily charged with ideologies, which is unfortunately not self-evident for many readers, especially young scholars. There are three particularly ideologized expressions that I deliberatively avoid in my writing: developing country, global South/North, poor (in the economic context or applied to a person; an argument can be poor).

The reasons for such avoidance are long enough to be left for another article.

Leave a Reply