A short(ish) explanation of why you were taught to write the way you were taught, the art we call Rhetoric…

This is an essay not about the fact that modern professional writing, especially on the Internet, is terrible, not detailing all the compositional sins of the modern journalist or blogger. Rather, this is an essay exploring why the writing fundamentals of previous generations were superior, why they work better—which, if you understand it, will be to your benefit in ways you don’t expect. Let’s leave behind, for a moment, all the existential threats to freedom, the battle for our national soul.  Let’s stipulate that a new Dark Age is upon us, Western Civilization already beyond the point of no return, and let us turn for a few minutes to the skills we will need, you and I, in order to preserve wisdom for some future generation, centuries hence, who, having clawed their way back from this abyss, will need to inherit our legacy if they are to have any hope of beginning the world’s next iterative experiment in liberty.  Plus, if you’re curious to know what bits and neurology and physics have to do with compositional style, you’re going to love this.

You have probably read many a complaint, on this site and others and especially in the hallowed digital halls of such institutions as National Review, on the decline of writing in modern American or English-speaking culture. You probably share these concerns. Open any random news article on the Associated Press website and witness the style of it: a barrage of one- or two-sentence statement-blocks, separated by paragraph breaks, which purport to constitute an “article.” This, to say nothing of the horrid baby-talk of Axios and similar. Reading it hurts your brain, though you might not be able to put your finger on why. “Time was,” National Review will tell us, “when journalists and other professional writers knew how to write.” When the quality of their prose was much greater. And the National Review columnist will wax long and eloquent on the subject, citing many very scholarly examples of how things used to be and drawing the parallel between this decline and the decline of our culture and society generally. (These are the writers, particularly at National Review, who make sure to write in a much higher voice—who, for instance, make sure to work in the word “shibboleth” at least once so that you know they are true intellectuals, to establish their literary fides if you will. “Fides” is another word they adore, along with any other Latin they can invoke. The longer and more obscure the aphorism, the better.  Fortasse sapientiam et exclusivam confundant.) But they never quite get around to explaining why, mechanically, the old way was better. It was. We know it was; we can feel in our bones that it was; but knowing that a thing was better, without knowing why or how, does not help us reclaim it. Were those long paragraphs and elegant sentences better just because they are artifacts of “Western culture” or “the Classics?” Or is it that the practices of classic Western culture were successful for objective underlying reasons? Do we love the writers of old because they invented something beautiful, but which was ultimately an arbitrary stylistic fad, doomed to pass away with time? Or had they, in fact, discovered something beautiful, something which is actually fundamental, universal, and timeless?  Something… correct, from the foundations of our universe, regardless of culture or era? Here is a primer on why writing evolved into that form which we had, and celebrated, and lost, and the absence of which we now lament.  It is explanation of why good writing is good and correct regardless of who uses it or lauds it and who doesn’t. Those choppy pseudoparagraphs of a typical Medium.com post or AP or Axios news story are our jumping-off point, and we’ll come back to them in a bit, but first we must get down to basics.

It is commonly asserted in information sciences that the most basic unit of information is the bit (short for “binary digit,” a single 1 or 0 in a binary number), but this is not quite correct.  It might be better to say that a bit is the most basic unit of data, but data and information are not quite the same. Rather, data are the stuff of which information are made.

A datum (singular of data) would be something like a number, or a word, or even a sentence or a color. A single something which can be represented using an alphabet or character set. A character is a single unit of an alphabet, and the simplest alphabet possible is binary, an alphabet of only two characters, which we conventionally call ‘0’ and ‘1’. Any more complex character-set I can represent using binary. For instance, I any number I can invent in the decimal system (an alphabet of ten characters) I can convert to binary.  116 in decimal would be 1110100 in binary. To any letter or other symbol in English I can simply assign a number—‘t’ we will say is letter 116 on our list of possible letters and symbols—and thus I can represent any letter or symbol in binary.  As long as you know you’re looking for English letters and symbols from that list, if you come across 01110100, you know that represents the letter ‘t’. (There’s the rub, but hold on for just a moment.) Even colors, if I just use a color-value table, can be represented in numbers, and therefore in binary numbers. Anything I can represent as a number I can represent with a binary number.

The one thing I cannot do is represent binary values by some even simpler system. The only thing simpler than binary would be unary, only a single symbol.  We imagine this would be a system of counting by making scratch marks.  Instead of counting 0, 1, 2, 3, 4, or 0, 1, 10, 11, 100, I would just make marks equal to the number, as earliest man might have done: |||| to represent four of whatever, four chickens, four slain enemies. But, this is not actually a unary alphabet. In fact, in practice, it becomes simply a more ungainly form of binary, because it still requires two symbols: my scratch mark, ‘|’, and blank space, so I know where my number begins and ends. A blank space is as much a character in your alphabet as the spacebar is a key on your keyboard. Without a blank space, I would just have an endless string of scratch marks, and you would never be able to derive any meaning from it. (There’s the rub again, but hold on….)

So, binary is the minimum requirement for a language of data, and any datum can be represented by some chunk of binary digits. Indeed, we can define data as anything which can be represented by a number, or a sequence of numbers, and therefore we can further define data as anything that can be represented as a sequence of bits. But, is data the same as information?

Yellow.

That’s a datum. You’ve just been given a unit of data. Have you been given any information, though? Have you gained an answer, or have you only gained questions? “Yellow what? What’s yellow? Is something yellow? Something about yellow? Do you want something yellow? Why are you saying yellow to me?”

16776960.

Same problem again. I have given you some data, but have I given you information?  Just shouting a number at you is the same as just shouting “Nee!” at you, or barking at you.  It’s just a sound—until you can place it in a context.  Until it answers some question you already had. Either that, or I must provide you the question and the answer at the same time.

16776960 is the color code, in decimal, for Yellow. (Not that I would ever give it to you as a single decimal value. I would give you 255-255-0, or FFFF00, but that’s another story.)

Here I have actually passed you some information, because I have given you one datum in th context of another. With context, you are able to assign meaning to the datum, and with meaning, a datum becomes information. Thus we get the true definition of information. A data scientist would say information is “a non-random sequence of bits,” but ask him what he means by that, and he might struggle to answer, so I shall answer for him. He means “a sequence of bits with meaning,” such that if you change any of the bits, you change the meaning. Another word for a sequence of bits is a datum, so information is data with meaning.

That’s the rub. When you say, “with meaning,” it’s like saying, “the universe,” a small phrase with big implications. In fact, the phrase “with meaning” is actually bigger than “the universe.” The difference between “a sequence of bits” and “a sequence of bits with meaning” is the difference between dust and a person, between no universe and a complete universe, between smudged paper and Shakespeare.  It is the difference between nothing and everything. As an aside, you will run into naturalists who say things like, “A thought is just a sequence of electrochemical discharges across the brain.”  You say, “No, a thought is a sequence of electrochemical discharges across the brain that has meaning.”  And they’ll say, “Well, sure,” as if that’s a trivial distinction, and you must reply, “No, no ‘well, sure.’ That’s everything. The ‘with meaning’ is the most important part of the sentence. Discounting that is like discounting the difference between a regular housecat and a housecat with an entire living galaxy hanging from its collar. The most important part of that sentence, the alpha and omega of all science and philosophy, is that phrase ‘with meaning.’”  Remember that, next time you’re talking to a naturalist.

So, information is data with meaning. But this means that information can never be comprised of a single datum. For me to give you information, I can give you a single datum only if you already have at least one other datum of context. For instance, I can say, “Table!” to you, but I’m only giving you information when I do that because I saw you hunting around the kitchen muttering, “Keys… keys… keys….” You already have a question, “Where are my keys?” which amounts to at least one datum, and I supply an answer, “On the table.” If you don’t already have at least one datum of context, then for me to give you any information, I must give you at least two data, one referring to the other.  “Keys → table.”  That’s enough for you to have some hope of gleaning some meaning from the data I passed, and therefore to have some hope of receiving the information it represents.  And even that much requires that we already share a common presumption that you might be looking for your keys, and that if I link the one datum, “keys” to the other datum, “table,” I’m making a statement about the spatial relationship of the two. The less information we already hold in common, the more data I’m going to have to send to you in order to communicate actual information, but the minimum number is two.  “Car!” or “Red!” in isolation are just barks, just noises, but “Car → Red!” at least has a chance to be information. You know, at a minimum, by this statement, that “car” and “red” have some relationship, and the more other data you have, the more you can guess at the nature of that relationship.  Information thus can be defined as the relationship between at least two data.

Now, that has a name, that conceptual structure we create when we link one datum to another datum. It’s called a “map.” You probably thought a map was the thing you pinch and swipe on your phone to see exactly where your Uber driver has been delayed, but that’s… not what you think it is.  A map is, in fact, just a list of connections between data, from one datum to at least one other datum.  The most basic map is a list of a single connection (or “link” if you prefer) between one datum and one other datum.

A → 1

More commonly, we tend to run into maps which are lists of linked pairs:

A → 1

B → 2

C → 3

These are very common…

Car → Red

Truck → Blue

Carla → 555-1212

Bob → 555-8725

But the point is, it is this linking process, linking a datum to another datum, which begins to build information out of data, and the minimum unit of information is, thus, a map of at least one link: between the datum you’re thinking about and the datum representing what you’re thinking about the first datum. That’s a confusing way to say it, so let’s put down some labels:  The definition of information, restated, is a map, and the minimum unit of information is a single-row, single-link map, mapping a single subject datum to a single predicate datum.

Subject → action

John → shouts

Spot → runs

I → am

I said earlier that the more data you include, the more likely you are to have encapsulated genuine information, and you can begin to see where this is going.  “John → shouts” is enough.  You have made a meaningful statement, in a way wholly and qualitatively different from if you just stood there barking “John!” over and over, with nothing else, or “Shouts!” over and over, with nothing else.  But if you give me a more complex map, “John → shouts → profanity,” now I’m really starting to build an image in my mind something like the image that you have in your mind. Now we’re building a complex predicate. We’re not just linking the subject to a single datum, but to sub-maps of multiple data. This is essential for passing information on state-of-being.  “Car → red” is barely information.  “Car → [being verb] → red,” though, is getting somewhere.  “The car is red.”  And you can add as many links as you like, complicating your subject, complicating your predicate.  “(John → [conjunction] ← Judy) → lock → lips ← [preposition] ← (seat ← back).”  If a list of maps of different subjects to different predicates, such as we had above, is a compound map, then this map with multiple links in a single row we might call a complex map. Compound and complex maps each contain greater quantities of information than a simple map, but each in a different way.  A compound map contains many instances or examples of one (usually fairly simple) kind of relationship, such as many instances of as simple “has” relationship.  Each [person] has [phone number].  Each [car] has [color]. A complex map, by contrast, describes a complex relationship between many data, as in the complex relationship between John, Judy, a kiss, a car seat, and the various actions and states of being involving these.

Okay, so, the basic unit of data is a bit, sure, but the basic unit of information is a map of at least one subject datum to at least one predicate datum.  Yes, you are correct; we’ve invented the sentence.  Or, rather, we’ve discovered the sentence, and that is really my point, here, today.  This is not something we created. We did not decide that the basic unit of information would be a map of at least two data. It is what it is. It is a fundamental truth of our reality, built into the fabric of our universe. Sentence structure is not arbitrary. The “sentence” is just the verbal representation of the basic unit of information as information exists in our universe. A compound map, with multiple rows, would be multiple sentences, or a compound sentence.  “The truck is blue, but the car is red.”  (Your paper or digital road map, by the way, is actually a very long compound map of simple relationships. It’s a list of points/pixels in a picture each linked to a single point/location in the real world.)  A complex map would make a complex sentence.  “John and Judy lock lips in the back seat.” When you learned in grammar school to write proper, complete sentences (and to diagram them, if you were lucky enough to get a good education), you were learning to handle and keep track of the links in a complex map.

Now, obviously, a single sentence, even an elegant, very complex sentence, is not going to get us very far.  Likewise, a simple compound map, even a very large one, will only serve to communicate a limited class of information.  Think about the examples above, of lists of customers and their phone numbers or cars and their colors.  These are the maps you typically encounter at work, usually on a computer as a database or spreadsheet or, in the old days, stored in a filing cabinet in the form of reams of customer records.  We have determined that all information can be represented as maps of various kinds, but what kind of map represents it, what form it takes if you visualize it as a map, will depend on the kind of information and your purpose in storing and conveying it.  We are concerned here not with the ability to store and convey ten thousand customer records, but rather with the ability to store and convey a single complex map—really, a single row map, but one which goes beyond linking one subject to one predicate, even if both are quite complex, as in our kissing example.  We’re talking about building a subject and predicate, and linking that entire thing to the next subject, and linking that to a predicate, and linking the results of that link to yet another subject and predicate, and so on, to create an ever increasingly complex train of logically-connected information.  We’re talking about a map that might only have a single row, but that row contains thousands or tens of thousands of links, far beyond the basic subject-predicate structure, nesting links within links and relationships within relationships.

If the proper name for long tables of relatively simple maps (as in customer records) is “database,” then the proper name for this data structure, a single but extremely extensive, extremely complex multi-sentence map, is “idea.”  Whether I am trying to explain something (to you or to myself, so that I know I fully understand it), or to convince you to believe what I believe, or to build a static image in your mind, or to convey to you (or to myself for better understanding) how a system changes over time, the common thread is that I am trying to encapsulate, in language, a complete idea, no matter how complex that idea.  I am trying to build an extremely complex map, a train of countless links between data, each building logically on the last, until the idea is complete.  If you have ever had an idea, you have contained such a map in your head, even if you did not recognize it.  And to the extent that you were able to bring your idea into focus, make it detailed and, ultimately, useful, something you could put into practice, is the extent to which you were able to map out that idea, usually by means of language.  Colloquially we call this process, “Thinking it through.”  In order to implement your idea, or communicate your idea, or even to be sure you really, completely had the idea at all, you had to think it through, completing the map from datum to datum, link by link.

Some ideas, it’s true, are not mapped out verbally.  Rembrandt mapped out his ideas as color on canvas, and Mozart as notes on musical staves.  (Mozart is a peculiar example. Uniquely among composers, he did not “sketch out” his compositions as rough notes and then go back and work out the details, but rather wrote out, in a single pass, entire complete orchestrations in their final form. His mind was such that it could contain the complete map, fully realized as notes on staves, and all he had to do was “copy” it onto staff paper as fast as he could move a pen.) Most ideas, though, held by most of us mortals, are of the kind best represented by the medium of verbal language.  That, too, has a name, that rendering of the idea-map into words of English (or Japanese or French or Latin).  Whereas in music we call it composing and in the visual arts we call it painting or sculpting or rendering, in language we call it rhetoric. To “think through” an idea using words, i.e. to fully map it out using words, is the definition of rhetoric.

(The philosophers of old recognized only four possible rhetorical purposes, which I mentioned earlier and which may have rung a bell in your memory.  We might better or more fundamentally call them the four types of rhetorical maps, but there remain only the four:  Description, to render the appearance of a system to the senses of an observer; Narration, to render how a system changes over time; Exposition, to explain the nature or underpinning logic of a system; and Persuasion, to render argument in favor of a proposition. Any time you engage in rhetoric, you will be mapping out an idea of one of these four kinds, and never more than one at a time.  You may perform one in service to another—for instance, describing a thing visually so that you may then describe how it changes over time—as part of a larger, overarching idea, but your rhetoric at any given moment will be to one, and only one, of these four purposes.)

Because an idea can be so complex, a single unit of information, a single sentence, may not suffice.  I may not have a single subject (even if it is a compound or complex subject) linked to a single predicate (even a compound or complex predicate).  Instead, I may have a subject linked to a predicate, and that whole linked to another subject, which links to another predicate, and several of those linked together in some way, and so on. I need to convey a lot of related information, each unit of information building on the last, toward one of those four goals.  So, I am going to need multiple sentences, but I am going to have to organize them in a way which preserves the links between them and supports the ever-expanding train of logical connection.

Here, I come to the necessaries of biology and neurology. The brain, the conscious mind, addresses ideas a certain way. First, it must orient toward or focus on the task at hand.  Then, so prepared, it can take in individual units of information, store them, and process them as related to one another and to the problem itself.  If they are meaningful and relevant to the problem, hopefully, the brain can synthesize from them a solution, a coherent new “understanding,” what we call a complete idea.  The minimum unit of information is a sentence, so we supply sentences to a brain (or the brain forms sentences for its own purposes) in this same format: one sentence to orient the mind, a few sentences to develop the idea, and a final sentence to synthesize a new understanding of a complete idea—or at least to let the mind know that the idea is complete, that that particular segment of communication has ended, and some synthesis should have occurred.  In a conversation or lecture (rhetoric as communication of an idea), this is typically where we pause for feedback.  Look for a nod of understanding, an “I get it” from the audience, as they achieve synthesis of the appropriate conclusion, the completed idea.

This structure, friends, is a paragraph.  A paragraph is not an arbitrary block of text, it is not however much you happen to say until you run out of gas.  It is not something you are supposed to write because your teacher told you to, but from which you are liberated once you are out of school.  A paragraph is one idea unit, transmitted through the medium of language, and it is absolutely vital to the communication of an idea to another person.  You are not psychic, so you can’t just move an idea into another person’s brain as a whole thing.  Like a Star Trek transporter bay, you have to break the idea down into atoms, convert those atoms into language, and send them one atom at a time, in hopes that the recipient will be able to rebuild the structure piece by piece, particle by particle.  One sentence to orient the target brain to the incoming transmission, several more to deliver the relevant atoms of information, and then one last to signal the end of the transmission and to guiding the target brain to synthesize the information into a complete idea—ideally, the idea as it looked in your mind, the idea you intended to convey.  Paragraphs are the lowest level of complete idea transmissible between brains—not because we choose it to be so, but because that’s how information works, and how brains work.  When your English teacher told you to write a proper topic sentence, supporting or developing sentences, and a concluding sentence, she was not just telling you what the old masters prefer, she was telling you how your universe works.  There’s no fighting it, any more than you can deny the reality of gravity or electromagnetism or genetics.  If you leave off any piece of this structure, you are in effect hoping that your recipient will be able to fill in the gaps, either to figure out for himself what the hell you are talking about (if you omitted a topic sentence to orient his mind), to interpolate your missing logical steps (if you omitted one of your supporting sentences), or to realize that you are done talking and to come to the correct conclusion (if you just trailed off without any kind of concluding/synthesis sentence).

The paragraph is the lowest level, the most basic unit of a complete transmissible idea, such that if you have not sent a full paragraph, you have not sent a whole idea, but only pieces of an idea and a fool’s hope.  Let us say that you do transmit complete paragraphs, though.  The paragraph is the minimum unit of idea, but it is not the upper limit on an idea.  Once the target brain has received a few paragraphs of information, a few discrete ideas, it can potentially synthesize those into a higher-level idea.  After building several higher-level ideas, it might even be able to synthesize those into a very-high-level understanding.

This process, too, reflects the format of information in our universe. Our universe has certain minimum elements which can’t be broken down any farther. I used the word “atom” earlier, and that is a better metaphor than you might think. The Ancient Greeks coined it to describe exactly that: the minimum element into which a thing can be broken down, but which itself cannot be broken down any further.  “Atom,” from “a- tomos,” literally means “not divisible.” An atom is the minimum component of everything else.  Once you have your atoms, everything else is built out of them level by level.  Combine atoms into molecules.  Combine molecules into substances.  Combine substances into objects.  Combine objects into systems.  You can break a house down into furniture, walls, floors, foundation.  You can break a wall down into bricks and mortar.  You can break a brick down into minerals, and break a mineral down into its constituent atoms.  The atoms, though, cannot be broken down into anything else.

“But wait!” you complain.  The atoms, we have since discovered, can be divided into subatomic particles.  Yes, that’s true, but there are only three subatomic particles of matter.  Any given atom can only be broken into a simple map of protons linked to neutrons and electrons.  As many different kinds of atoms as there are, they are only divisible into protons, neutrons, and electrons, and if you break an atom into its protons, neutrons, and electrons, you retain no information about the atom.  Protons, neutrons, and electrons by themselves don’t tell you anything about the atoms they were once a part of.  One proton is like any other and does not imply the myriad properties of the elements. Likewise, the basic unit of information, the simplest sentence or map, is divisible, but only into units of data which by themselves have no meaning.  As we discussed at the beginning of this analysis, if I break a sentence down into its subatomic particles, the individual words, and just give you one of those words, or even all of those words but without the links between them, I have given you no information.  To have information at all, you must start with that indivisible minimum of information, the sentence.  Once you have that, though, the rest is a process of building grander structures out of lesser structures.  You use a collection of sentences which convey an idea according to the way the brain handles an idea, the orient-develop-synthesize process we discussed above.  That’s a paragraph.  To convey a higher-level idea, you’ll use several paragraphs to send the component ideas.  However, you don’t want to just attack the audience with supporting paragraph-ideas, any more than you can just attack them with supporting sentences without first orienting them to what idea you are developing or later helping them synthesize the complete idea from those supporting sentences.  So, if your intent is to provide multiple paragraphs of information all of which support some grander idea, what will you do first?  You’ll create a whole orienting paragraph:

“Friends, I’m about to convey to you a complex idea called George.”—orienting sentence.  “It will be comprised of several smaller ideas conveyed as paragraphs.  First, there will be a paragraph for idea A, then a paragraph for idea B, then a paragraph for idea C.  Once I’ve conveyed all of those component ideas, I’ll show you how they combine into this overarching complex idea.”—developing sentences.  “If there are no questions, I’ll begin transmitting component idea A by moving on to that paragraph.”—concluding/synthesizing sentence.

Then you’ll send paragraphs for each of your component ideas, orienting the audience to each, developing each, synthesizing each and signaling the transition to the next.  Finally, you’ll wrap up your transmission with a whole paragraph designed to help your audience synthesize the lesser ideas into the greater:

“So, you have all three component ideas.  Let’s synthesize them into a complete understanding of the problem.”—orienting the audience to the problem of overall synthesis.  “You can see that idea A supports idea B, and you can see that idea B supports idea C.”—supporting sentences, calling back to the component ideas.  “Well, George must be true if C is true, and A and B are known to be true, and A and B together imply C; ergo, George!”

“By George, I think he’s got it!” they all cry, as the light of greater understanding dawns.

An essay is a collection of paragraphs which has this structure.  As a paragraph has a topic sentence, developing sentences, and a synthesizing sentence, so an essay has a topic paragraph, developing paragraphs, and a synthesizing paragraph, like the examples given above, or the example which follows.  In this one, you are a computer science professor, and you have assigned your students to write a simple mathematical program for a very simple computer processor which you helped them design and prototype.  One of your students comes to you for help.

“Professor,” says he, “I’m trying to write this program to solve this problem you assigned on our prototype computer processor, but our processor only performs adding, and I’ve thought it through every possible way, but this problem requires that we perform a division operation.  I don’t see how it can be done.”

“Okay, well, consider this,” you say, orienting the target brain, your student’s brain, to a new idea. “1 + 1 = 2.  And 2 + 2 = 4.  3 + 3 = 6.  4 + 4 = 8.”  You thus convey several units of information, several sentences. “You see?  The doubles of your counting numbers are your even numbers. 1, 2, 3, 4; 2, 4, 6, 8.”  Synthesis, for a very simple idea.

“Yeah, okay, that’s great,” says the target brain, “but why do I care?”

“Because, check this out.” Orienting. “If I want to divide any number by two, all I have to do is count my evens, and see how long it takes me to get there.  So, sixteen divided by two: 2, 4, 6, 8, 10, 12, 14, 16.  It took me eight numbers, eight increments of two, to get there.  So, sixteen divided by two is eight!”  Development, through another example, and synthesis, the relation of counting by twos to division by two.

“Huh.  Okay,” he says. “I never thought of it that way, but I get it.  I still don’t see how this helps me, though. This problem isn’t limited to even numbers.  I need a generic solution.”

“Well, let’s try the same thing with a different increment.” Orient.  “Can you count by threes?  3, 6, 9, 12?” Developing.

“Yeah.”

“So how many was that?  To get to 12?” Developing.

“Four… Oh.  Twelve divided by three is four. I get it.”  Synthesis.

“So, can you, in fact, perform any division of any number by any other number, using nothing but addition operations?”  Orient the student to the final idea–

“Ooooh,” says your student.  “Yes!  You can!  You just increment by the divisor and count the increments!  It’s just a loop, with two separate addition operations.  The increment, and the counter.  Oh, man, that’s wild.  Thanks, prof!”

Did you see the process playing out?  Where was the introductory paragraph in this essay?  It was, in fact, provided by the student.  He orients the conversation to the problem, develops the problem, and synthesizes a thesis: the problem cannot be solved—unless the professor can offer some secret new knowledge.  The professor responds with developing paragraphs, which I labeled for you.  At the end, the professor begins the concluding paragraph, orienting the conversation toward the task of synthesizing the overarching, higher-level idea, using a rhetorical question, and the student takes over the concluding paragraph from there, as understanding comes to him. He describes the generic solution with developing sentences and then concludes that paragraph with an expression of gratitude.

(Actually, if you’re interested, division is probably better described as a series of subtractions, but I’ll let you think about that on your own.)

If you are developing an even grander idea, then you might have multiple essays over the course of a chapter of your book, and the chapter of your book might begin with a short essay of several paragraphs explaining what the purpose of the chapter is and what specific sub-sections will be developed in it, and that chapter will probably conclude with an essay of several paragraphs synthesizing everything that chapter covered.  Your book, meanwhile, will have a prologue or introductory essay, maybe even a complete introductory chapter, and a concluding essay or chapter as well.

I want to draw your attention to a peculiar thing that is happening here.  Let us say I simply labeled parts as, “Orientation,” “Development,” “Synthesis.”  And I told you, “Here is an idea with an orientation, a development, and a synthesis,” could you guess, from that description, whether I was talking about a paragraph, an essay, a chapter, or the whole book?  Nope.  Each level has exactly the same structure.  My third development chapter breaks into an orientation essay, seven development essays, and a synthesis essay, and each of those essays contains orientation, development, and synthesis paragraphs, and each of those contains orientation, development, and synthesis sentences.  If I just describe the structure, the shape of it, and nothing else, you have no idea how “zoomed in” we are, whether we’re zoomed in on a single paragraph or zoomed out looking at the whole book, or some level in between.  It’s a bit like a tree.  If all you could see was a piece of its shape, without any of the details of texture or foliage, could you tell if you were looking at the trunk dividing into main boughs, or a small branch dividing into twigs and stems?  There’s a word for this, folks, this concept that a pattern repeats not in sequence but in subdivision.  It’s called a fractal, and it is one of the dominant mathematical phenomena in nature.  From trees branches and tree roots, to snowflakes, to coastal fjords, to watershed or river-delta topography, to the golden ratio, systems in nature compound fractally from their atomic components to their upper bound of growth.  There is an atomic unit of water runoff, a certain minimum of water necessary on a given earthen surface before it joins and forms into a rivulet, but from there rivulets join into rills, rills into brooks, brooks into streams, streams into rivers, rivers into greater rivers (the White Nile and Blue Nile into the Nile proper), a pattern repeating not in sequence but by multiplicative expansion until the river falls off the edge of a continent and runs into the sea.

Communicated information, being a natural thing, has this same structure.  As a branched structure has its atomic unit (the line segment) and the fractal pattern by which it grows (segments branching into other segments), so rhetoric has its atomic unit, the sentence, and then the fractal pattern by which it grows out of its atomic units: orient section, development section, synthesis section.  Fractal expansion is, again, as fundamental to our universe as gravity, electromagnetism, genetics, and the dual nature of man.  When your English teacher taught you about introductions and developments and conclusions, she was not just describing The Way It’s Properly Done, according to some admittedly brilliant but now dead and no longer authoritative old British or Greek writers.  She was describing The Way It Is, such that you can embrace it, and convey your ideas, or you can ignore it, and never communicate (or even understand) your ideas quite as well as you would like.

Now, there are ways to play with the structure.  A fractal pattern of this kind has a rhythm to it.  It is musical.  If I build it and build it and build it, so that you are accustomed to it, then when I suddenly interrupt it, you will feel that interruption of rhythm, and it will draw your attention, just as if I had hit a sudden all-piece rest in a song, and then a singular, staccato chord.

Here’s an example.

You notice that one-sentence paragraph, do you not? It’s not complete.  It’s not, technically, a paragraph at all, at least on paper. It serves its purpose, though.  It even conveys a paragraph’s worth of meaning.  The next component idea I wanted to convey to you was an example.  I wanted to orient you that I was about to present an example, demonstrate the example, and then let know that the example was complete and synthesize the properties it demonstrates.  I could have said, “Let me give you an example of an interrupted rhythm.  Imagine if, after the previous paragraph, I had simply typed, ‘Here’s an example,’ with a paragraph break before and after. It’s one sentence. Not in any meaningful sense a paragraph at all. By marking it out as a paragraph but leaving it incomplete, I have broken the paragraph rhythm.”  I could have said that, but it seemed to me much more effective simply to demonstrate, without any wrapping.  Does it not have more impact that way?  And it still serves the purpose of a complete paragraph about an example.  You don’t need an orienting sentence, you don’t need me to explain the example or tell you that I’m done giving the example.  The example is enough by itself.  Why?

Because you’re not an idiot.  If I slam on the brakes just there and simply type, “Here’s an example,” with a paragraph boundary before and after, you’re intelligent enough to know that it is self-referential, to figure out from the context of the moment what it is an example of, and to pick up in the next real paragraph that I am discussing that example, without the need for a transition sentence.  I can be incomplete because I can rely on our already-shared knowledge, our already-shared perspective.

This is the only way it works, folks.  If I do not give you complete paragraphs at the paragraph level, and complete essays at the essay level (with orienting, developing, and synthesizing paragraphs), and complete chapters at the chapter level (with orienting, developing, and synthesizing essays), and so on—if I do not make the fractal pattern complete at each level of the work in question, then I am not conveying complete ideas.  There’s no getting around it.  I can deviate, occasionally, playfully, in situations where I know we already have enough shared knowledge that you will be able to fill in the gaps, but aside from that, I must honor the natural structure of information as God has made it, or I will fail in my rhetorical goal.  I can do everything right up to the level of the chapter, but if my book does not have a good intro and conclusion, you will come away feeling like I was trying to make a point but you’re not sure which one.  I can have a good opening and conclusion to my book, but if the individual essays do not orient you to their subjects, develop their subjects, and help you to synthesize those subjects into coherent supporting ideas, then you will get to the conclusion of my book and ask, “Wait, when did he say that? Was that supposed to be somewhere in the jumble of the previous 500 pages?”

It’s worth mentioning at this point that most intercourse (you there, get your mind out of the gutter) is not rhetoric.  Most intercourse in the daily life of man or woman does not and is not intended to convey ideas.  Most intercourse takes a form which might be exemplified by this example:

Me: “Man, those liberals, am I right?”

You: “Yeah, man, you know it.”

Orientation: “Man, those liberals…”  Synthesis: “…am I right?”

Synthesis: “Yeah, man, you know it.”

There are no ideas here. There is no communication of information. The initiator did not even send a complete sentence.  What’s happening, here?  My earlier example of an interrupted rhythm, I said, depended on pre-shared knowledge.  This is that taken to the highest degree.  That was me relying on pre-shared knowledge to create a (hopefully) artistic and illustrative interruption of rhythm in an otherwise complete rhetorical piece which (hopefully) enhances its rhetorical effect.  This is… nothing.  There’s no context except what we already share, our presumedly common regard for “those liberals.”  There is no proposed mission of rhetoric, no intent at all to transmit ideas.  Why did we even, biologically, spend energy on this exchange?

This is what transactional psychologists would call a transaction, probably worth about three strokes each.  I call it barking, or in this case maybe nose-sniffing.  Our intent is not to communicate but to acknowledge one another and reinforce tribal bonds.  It has the form of human words, but it is a purely animal interaction, equivalent to two dogs who know each other passing one another on their respective daily walks.  We aren’t strangers, so we don’t need to go through that whole ritual of tentative anal sniffing or aggressive snarls and other challenges, but neither are we of the same household, so we haven’t seen each other since yesterday or the day before, and some greeting is demanded to ensure our mutual nonaggression and neighbor status.  The appropriate interaction, then, is a quick nose-to-nose sniff between neighbor-dogs before each of us is dragged away by our respective owners. Most of what comes out of the mouths of human beings is this.  Yes, again, it has the form of language.  I know, sometimes it even comes in complete sentences.  Between you and your besty, it might go on for hours, neither of you shutting up but long enough to breathe on occasion.  Even so, most of it could be replaced with dog sounds and dog behaviors and it would lose none of its content.

(The other possibility, since it’s me and you in this example, and I know you’ve read this article, is that I’m using a bark about liberals to imply a bark about barking, as a kind of self-referential inside metajoke which I know you will get because you have read this article. If you got the shibboleth joke way back at the beginning, then you’re probably down for that sort of thing. If not, then we may think we’re barking about the same thing when in fact we’re barking about two entirely different things, each laughing at an entirely different joke.)

To reiterate, then, these are your options:  either you are engaged in rhetoric, transmitting ideas properly formatted, or you are engaged in barking, transmitting no information but only transactional strokes dependent entirely on common tribal membership.

Actually, there is one other possibility: that you think you’re doing one thing, but actually you’re doing the other. And that brings us back to the example I used to kick off this adventure, that specific literary crime of modern journalism in which an “article” is comprised of nothing more than a machinegunning of stubby pseudoparagraphs containing isolated declarative statements. Each of these pseudoparagraphs contains only one or two declarative sentences, with no orientation or synthesis, so none of them is actually a paragraph.  So, what’s going on in these articles?  Charitably, we might theorize that this person just doesn’t know when to hit the Return key (that’s the Enter key except when you’re writing) and when not to, and so has put in extra, specious paragraph breaks.  By this theory, there is a paragraph there; there is an orienting sentence, followed by several developmental sentences, followed by a sentence of synthesis.  All we need to do is remove the extra whitespace, pack the appropriate sentences back together spatially, and there the proper paragraph will be.  Is that what’s going on?  Let’s take another look at one of those AP articles.

Nope.

There is, at best, something like an orienting sentence at the beginning of the article, but usually not even that.  You get instead just the article title or headline, which seems to be trying to do that job, and you might get section titles as well, as in an Axios piece.  “What it all means.”  “Why it’s important.”  Followed, inevitably, by more isolated declarations.  No, there is definitely no paragraph structure here at all, nor any corresponding organizational structure at the higher fractal levels of the composition.

Yet this is not just a friendly, informal barking exercise, or at least it does not seem to think of itself that way.  This person clearly thinks he or she is engaged in expository or persuasive writing, or, in the case of a reporter, a narrative account of events.  Otherwise, why bludgeon us repeatedly with these declarative statements?  These are clearly units of information (even if most of them later turn out to be false) which might have been used to develop an idea, but, with no orientation or synthesis to bind them, they just sit there on the page.  It’s a bit like a small, preverbal child who has walked up to you and set his toys down in front of you.  Here are the pieces.  He has an intent for you.  He’s going to stare at you and hope you figure it out.  It’s a bit like your dog, dropping a ball at your feet and then staring at you, maybe bouncing in circles a few times, and then staring at you again.

This is the only conclusion we can draw from what’s before us. This person thinks he is engaged in rhetoric, but he has no understanding of rhetoric, so he is making a brutish, grunting imitation of it, spewing statements onto a page and hoping, by the end of it, that people who are already members of his tribe will glean some vague semblance of the idea that was in his head, and that people who are not members of his tribe will go die in a fire. It is unfortunate, but it is also somewhat inevitable.  It is inevitable that a person who believes nonsensical things should write this way.  And this is the surprise twist, here at the end.  Let me explain:

A person who writes this way does not write this way because he never learned to write.  He writes this way because he never learned to think.  Proper composition is not the product of a classical Western education; it is a product of the nature of, and natural structure of information in, this universe which God has made, which Western classicists observed, illuminated, and harnessed.  All of the above is not just a way to write well but the only way to write well, because it’s the only way to think well.  It is the only method by which your conscious mind can grapple with and organize complex ideas and synthesize them into even more complex ideas.  If you can’t do this on paper, it’s because you can’t do it in your head, and if you can’t do it in your head, then complex ideas will never be within your grasp.  They will always escape you.  Likewise, if you practice doing this on paper, you are practicing doing it in your mind.  By writing complex ideas (or speaking them) correctly, by properly representing their structure in rhetorical form, you are teaching yourself to think through complex ideas correctly, to grasp them and incorporate them.

The machinegunning of stubby pseudoparagraphs is just one prominent example of the decline of composition and the decline of thinking which it reflects.  There are other sins, and if you think back to your English classes, you’ll probably be able to identify them.  (For instance, being not well-trained in composition, the dangling participle is another common mistake of the typical Internet columnist.)  But your task is not to go out and identify all the basic grammar and composition mistakes they are making.  Your task is to go back to the basics yourself, hopefully inspired by this article, and review your High School English lessons, and think about how they were not teaching you to write, but teaching you to think.  If you can think, if you can lay forth complete ideas rhetorically, then you can preserve them for posterity, not as incomplete sketches which rely for understanding on common cultural reference and shared inheritance, but as complete, self-contained records which a future generation, born of another lineage and another culture entirely disconnected from yours, may discover and reconstruct into useful wisdom.  This was the gift of the Greek philosophers to the students of the Medieval age, that these descendants of Gaulish and Germanic barbarians, on the far side of a Dark Age in which all communication from progenitor to posterity was interrupted, who inherited nothing from their fathers and from their cultures but rare access to the Latin vulgate, were able to rediscover the works of Plato and Aristotle, whole, complete, ready for use, and to from these rebuild civilization and resume the progress of Western Civilization.  With the same skill, hopefully, you and I can leave behind artifacts which will outlast not just ourselves and our generations but our very civilization which is now in its death throes, carrying complete ideas down through time to whatever new civilization grows up in the ruins, providing those people, so long as they can at least decode the English language, not just hints of what came before, but a recipe for how to go ahead, greater and wiser than we were.