Communicating Clearly: Science and Practice of Clarity

In 2003, a senior engineer at a large aerospace contractor wrote a twelve-page technical memo recommending that the foam strike on the Columbia Space Shuttle's thermal protection system warranted immediate investigation.

The memo was precise, detailed, and organized around the sequence of his thinking: background, data collection, preliminary analysis, concerns, further data needed.

The recommendation was buried on page ten. His supervisors, reading quickly under deadline pressure, concluded the memo indicated the strike was not a serious concern.

They were wrong. Columbia disintegrated on re-entry two weeks later, killing all seven crew members. A post-accident investigation found that the communication structure of the memo was a contributing factor: critical uncertainty had been framed as technical detail rather than as the central message.

This is a catastrophic example of a failure pattern that plays out in offices, hospitals, courtrooms, and relationships every day at a scale that is merely ruinous rather than fatal. The engineer knew something important. He communicated it in a form that buried the importance.

The gap between what he knew and what his audience understood was not a knowledge gap or an intelligence gap. It was a communication structure gap.

Research on communication failure consistently implicates structure rather than vocabulary as the primary culprit. The curse of knowledge, cognitive load, the wrong sequencing of information: these are the mechanisms behind most unclear communication. Understanding them is the foundation for fixing them.

"The great enemy of clear language is insincerity. When there is a gap between one's real and one's declared aims, one turns as it were instinctively to long words and exhausted idioms, like a cuttlefish squirting out ink." - George Orwell, Politics and the English Language (1946)

Key Definitions

Curse of knowledge: A cognitive bias identified and named by economists Colin Camerer, George Loewenstein, and Martin Weber in a 1989 paper, and popularized by Chip Heath and Dan Heath in Made to Stick (2007).

Once you know something well, it is genuinely difficult to reconstruct ignorance of it and communicate in a way that accounts for what your audience does not know.

Cognitive load: The total amount of mental effort being used in working memory at any given moment, a concept introduced by educational psychologist John Sweller in 1988.^[2]

Communication that generates high cognitive load, through dense jargon, complex sentence structure, or information presented in a non-intuitive order, exceeds the audience's processing capacity and reduces comprehension.

Pyramid Principle: A communication framework developed by Barbara Minto at McKinsey & Company in the 1970s and published in her 1987 book.^[3]

The principle holds that ideas should be communicated top-down, with the conclusion or recommendation first, followed by supporting arguments, then evidence, rather than bottom-up (reasoning first, conclusion last).

SCQA framework: Situation, Complication, Question, Answer. A narrative structure derived from the Pyramid Principle that is particularly effective for persuasive written communication. Establish what is currently true, introduce what has changed or is wrong, articulate the question this creates, then deliver the answer.

Plain language: A communication approach that prioritizes the audience's comprehension over the writer's expertise display, characterized by shorter sentences, active voice, common vocabulary, and clear structure.

The US federal plain language initiative, codified in the Plain Writing Act of 2010, produced comprehension improvements of 30 percent or more in controlled studies of government documents.

Why Communication Fails: The Curse of Knowledge

In 1990, a Stanford doctoral student named Elizabeth Newton designed an experiment for her dissertation in which one group of participants, the tappers, were asked to tap out the rhythm of a well-known song on a table while another group, the listeners, tried to identify it.

Before the experiment, the tappers predicted that the listeners would identify the song about 50 percent of the time. The actual success rate was 2.5 percent.

The tappers were hearing the song in their heads as they tapped. They could not stop hearing it. The listeners heard only a sequence of irregular taps without the melody that made the rhythm meaningful. The tappers consistently failed to model what the experience was like without the song.

This experiment is the empirical heart of the curse of knowledge. Chip Heath and Dan Heath used it as a central illustration in Made to Stick (2007), arguing that expertise creates a systematic communication liability: the expert knows so much, so automatically, that the experience of not knowing has become genuinely inaccessible.

Technical writers, scientists, lawyers, and specialists of every kind routinely omit context that feels obvious to them because they cannot reconstruct what it feels like to encounter the material for the first time.

The curse of knowledge explains why expert communication is so often opaque. It explains why academic writing is difficult even when the underlying ideas are not.

It explains why the engineer's Columbia memo was organized around his analytical process rather than around his audience's decision. He wrote the memo as a record of his thinking, not as a tool for his audience's action.

The corrective requires active audience modeling. Before writing or speaking, ask: what does this person already know? What is the decision or action they need to take? What is the minimum they need to understand to take that action?

The answers to these questions should determine what you include, what you exclude, and what order you use.

John Sweller and Cognitive Load Theory

John Sweller introduced cognitive load theory in a 1988 paper in Cognitive Science, arguing that instructional design must account for the limited capacity of human working memory.

Working memory can hold approximately four items simultaneously (later research by Nelson Cowan refined this from the earlier estimate of seven), and can process only so much novel information in a given time window.

Communication that simultaneously demands processing of unfamiliar structure, complex vocabulary, and dense logical reasoning exceeds this capacity before the message lands. The audience is not unintelligent; their working memory is simply full.

Cognitive load theory divides mental effort into three types: intrinsic load (the genuine complexity of the content), extraneous load (complexity introduced by poor presentation), and germane load (effort that contributes to learning).

Effective communication minimizes extraneous load without reducing the intrinsic complexity of the content. This is done through familiar structure, plain vocabulary, chunked information, and sequencing that builds on what the audience already knows.

A document that uses bullet points to present information that has a natural sequential relationship actually increases cognitive load by destroying the connective tissue that aids comprehension.^[10]

Conversely, a document that presents three-level hierarchies of nested clauses with technical terms in every sentence imposes extraneous load that the reader's working memory has no room to accommodate.

Research on readability, beginning with Rudolf Flesch's pioneering work in the 1940s and formalized in the Flesch-Kincaid grade-level formula developed in 1975 for the US Navy, consistently shows that shorter sentences and simpler vocabulary correlate with comprehension across audience types, including highly educated readers.

The assumption that sophisticated audiences prefer sophisticated prose is not supported by comprehension research; they prefer clarity, which requires more discipline than complexity.^[9]

George Orwell's Six Rules

George Orwell's 1946 essay Politics and the English Language remains one of the most cited and influential documents in the history of writing instruction, not because it is the most scholarly but because its rules are specific, memorable, and backed by subsequent empirical research on readability.^[1]

Orwell's six rules are:

1. Never use a metaphor, simile, or other figure of speech which you are used to seeing in print.
2. Never use a long word where a short one will do.
3. If it is possible to cut a word, always cut it.
4. Never use the passive where you can use the active.
5. Never use a foreign phrase, a scientific word, or a jargon word if you can think of an everyday equivalent.
6. Break any of these rules sooner than say anything outright barbarous.

The sixth rule is the most important, and the most often omitted when the rules are quoted. Orwell was not prescribing mechanical simplicity; he was arguing for clarity as a discipline, with judgment as the override.

His target was not complexity per se but what he called the invasion of one's mind by ready-made phrases: bureaucratic abstractions, political euphemisms, and technical jargon that substitute the appearance of meaning for meaning itself.

Research on active versus passive voice bears him out. A 2014 study by Gauger and colleagues found that active voice sentences were comprehended significantly faster than passive voice sentences at comparable levels of sentence complexity, and were rated as easier to understand by readers across educational levels.

Plain language research from the US government's own studies found that rewriting government notices from passive, bureaucratic prose to plain, active prose improved citizen comprehension from around 40 percent to over 70 percent.

Barbara Minto and the Pyramid Principle

Barbara Minto joined McKinsey & Company in 1963 as one of the firm's first female consultants and spent years developing a framework for structuring consulting documents and presentations.

The Pyramid Principle, which she published in 1987 and has been teaching to McKinsey associates ever since, addresses the fundamental mismatch between how we develop ideas and how audiences receive them.

We develop ideas bottom-up. We gather data, notice patterns, form hypotheses, test them, and eventually reach conclusions. This is the natural order of inquiry, and it is the order in which most people write and present.

The problem is that audiences receive information in a completely different psychological mode: they are trying to build a mental model of what you are telling them and why it matters.

Without knowing where the communication is going, they cannot organize incoming information into a coherent structure.

They arrive at the conclusion, if they read that far, already burdened by a sequence of facts and arguments whose significance they could not assess along the way.

Minto's pyramid reverses this. Start with the single governing answer or recommendation, the top of the pyramid. Below it, present the three to five supporting arguments that justify it. Below each argument, present the evidence or data that supports it.

Every level of the pyramid consists of ideas that are mutually exclusive, collectively exhaustive (the MECE principle), and grouped by the logic that connects them.

The SCQA framework is the narrative adaptation. Situation establishes the agreed context: what is currently true that both parties accept. Complication introduces the change or problem: what has disrupted the situation. Question is the implicit or explicit question the complication creates.

Answer is the response to that question, which then unfolds through the pyramid structure.

The most common failure mode in business communication is beginning with situation and complication and either never reaching the answer or arriving at it exhausted and underemphasized after extensive preamble.

The Pyramid Principle requires a discipline of clarity about what you are actually recommending before you begin writing, because you cannot start at the top of the pyramid if you do not know what the top is.

The Plain Language Movement

The US federal plain language initiative has a history stretching back to the 1970s, but it was codified in the Plain Writing Act of 2010, which required federal agencies to write all documents, including regulations, letters, and forms, in plain language that the public can understand and use.^[11]

The evidence for the effectiveness of plain language is substantial. Research by Siegel and Etzkorn at the US Social Security Administration found that rewriting benefit notices in plain language reduced the number of follow-up phone calls asking for clarification by 25 to 30 percent.

Studies of medical consent forms consistently show that plain language versions produce substantially higher comprehension among patients, with one study finding comprehension improving from 40 percent to 78 percent after plain language revision.

The resistance to plain language in professional contexts is mostly sociological rather than logical. Technical jargon signals membership in a professional community and serves real in-group communication functions.

The problem arises when in-group language is used for out-group audiences, which happens constantly in medicine, law, finance, and technology.

The medical profession's persistent use of Latin terminology and passive-voice constructions in patient communication is a documented contributor to non-adherence with treatment, because patients do not understand what they are being told to do or why.

The UK Government Digital Service, which redesigned government web content using plain language principles beginning around 2012, found that redesigned pages improved task completion rates by 50 percent or more in user testing, and reduced the number of citizens calling helplines for assistance with tasks the website was supposed to enable.

Edward Tufte and Visual Communication

Edward Tufte's work on data visualization, particularly The Visual Display of Quantitative Information (1983) and Envisioning Information (1990), provides the visual equivalent of Orwell's prose rules.^[5]

His central concept is the data-ink ratio: the proportion of a graphic's ink used to convey data rather than decorative or redundant elements.

Good visualizations maximize this ratio, removing chartjunk, including unnecessary gridlines, background colors, three-dimensional effects, and decorative elements, that consume attention without conveying information.

Tufte's analysis of the Challenger disaster is particularly apt in the context of this article's opening.^[6] Engineers at Morton Thiokol had data showing that O-ring erosion increased at lower temperatures. They failed to communicate this clearly to NASA decision-makers in the hours before the launch.

Tufte's analysis of their charts, published in Visual Explanations (1997), showed that the visualizations they used buried the temperature-erosion relationship under a confusing arrangement of variables and failed to present the complete data.

Had the data been presented in the simplest possible chart, temperature on the horizontal axis, O-ring damage on the vertical axis, the relationship would have been immediately visible. Clarity in communication is not merely an aesthetic value; it has life-or-death consequences in high-stakes decision contexts.

Chip Heath's SUCCES Model

In Made to Stick (2007), Chip Heath and Dan Heath analyzed why some ideas persist in memory and culture while others, equally true and apparently well-communicated, disappear.^[4]

Their analysis produced the SUCCES model (they removed a 'S' deliberately to illustrate that you can break your own rules): Simplicity, Unexpectedness, Concreteness, Credibility, Emotional, and Story.

The most practically useful element for communication clarity is concreteness. Abstract ideas are processed by different cognitive systems than concrete ones and are consistently harder to remember and act on. The advice "be more efficient" is abstract and produces very little behavioral change.

The advice "before sending an email, cut one sentence" is concrete and produces measurable change. The difference is not vocabulary but the degree to which the instruction can be directly translated into action.

Credibility in the Heath framework includes what they call "testable credentials": giving the audience the means to verify a claim for themselves rather than asking them to trust your authority. This is particularly important in professional communication where the audience is skeptical.

Rather than asserting that your approach will save money, showing the calculation that produces the number invites the audience into the verification rather than demanding acceptance.

Marshall Rosenberg and Nonviolent Communication

While most communication frameworks focus on the transmission of information, Marshall Rosenberg's nonviolent communication (NVC) model, developed in the 1960s and published in Nonviolent Communication: A Language of Life (2003), addresses the dimension of communication that most directly affects relationships and conflict: the expression and reception of needs and feelings.

Rosenberg's framework distinguishes between observations (factual, specific, time-bound descriptions of what happened) and evaluations (judgments about what the behavior means or reveals about the person).^[7] "You interrupted me three times in the meeting" is an observation.

"You are disrespectful" is an evaluation. Observations allow the recipient to hear and respond to the underlying concern; evaluations typically trigger defensiveness and escalate conflict.

The NVC four-part structure, observation, feeling, need, request, provides a template for difficult conversations that research on conflict communication validates.

Studies of marital communication by John Gottman at the University of Washington, published across several decades of work, find that the criticism-defensiveness-contempt cycle is the primary predictor of relationship failure, and that specific behavioral observation and "I" statements are the most effective patterns for addressing conflict without triggering escalation.

Michael Nichols and the Listening Gap

Most writing on communication clarity focuses on speaking and writing. Michael Nichols, in The Lost Art of Listening (1995), makes the case that the quality of listening determines whether communication succeeds regardless of the clarity of the transmission.^[8]

The poor listener is not usually passive. The poor listener is busy formulating a response, half-attending while planning what they will say next, filtering for evidence of threat or agreement, and missing the nuance of what is actually being said.

In professional contexts, the pressure to appear decisive and knowledgeable creates strong incentives to respond quickly rather than to listen fully, at exactly the cost of missing the information that would make the response actually useful.

Nichols identifies several listening failures that are common in organizations: listening only to the factual content of a message and missing the emotional signal; interrupting before the speaker has reached the actual point; listening for confirmation of what you already believe and filtering out disconfirming information.

The active listening techniques that communication trainers prescribe, reflecting back, asking clarifying questions, checking understanding before responding, are not sophisticated psychological interventions. They are corrections for predictable and structurally motivated failures.

The $37 Billion Problem

The Holmes Corporation's oft-cited estimate that poor workplace communication costs US businesses approximately $37 billion per year in lost productivity was developed through survey methodology and should be treated as an order-of-magnitude estimate rather than a precise figure.

What it captures is directionally real: when communication fails in organizations, the costs are not confined to the individual interaction. They cascade through rework, misaligned decisions, failed projects, customer complaints, and the slow erosion of trust that makes all subsequent communication less efficient.

Email specifically concentrates these costs. A McKinsey Global Institute report from 2012 estimated that workers spend approximately 28 percent of their working week managing email.

Much of this time is spent on clarification loops: a message sent ambiguously, a response seeking clarification, a clarification that generates further questions.

The Pyramid Principle applied to email, leading with the request or key information rather than burying it after extensive context, would eliminate a large proportion of these loops by giving the recipient what they need on the first reading rather than the third.

Practical Takeaways

Start with the conclusion. In almost every professional communication context, the audience is better served by knowing the answer first. Structure emails, reports, and presentations top-down: recommendation or key finding first, supporting evidence second.

The audience can stop reading when they have enough; they cannot retrieve the buried conclusion they never reached.

Model your audience explicitly. Before writing, answer: what does this person already know about this topic? What decision do they need to make or action do they need to take? What is the minimum they need to understand to do that? Let the answers determine what you include and exclude.

Cut by 30 percent. Most professional writing is too long by roughly a third. Words that do not do work create extraneous cognitive load. The discipline of cutting, after drafting, is one of the highest-value editorial actions available.

Use active voice by default. Passive voice is not always wrong, but defaulting to it, as most professional writers do, consistently produces longer sentences and more obscure agency. "We need to improve the process" is clearer than "improvements to the process are required."

Replace abstractions with specifics. Every abstract recommendation in your communication should be replaceable by a concrete behavior. "Be more proactive" is not a communication; "send a weekly status update every Friday by 5pm" is.

Apply the SCQA framework to difficult messages. For any communication where the audience needs to be persuaded rather than simply informed, structure the message: what is the situation we agree on, what has changed or is wrong, what question does this create, what is the answer.

Listen to understand before listening to respond. The quality of your communication is bounded by the quality of your listening. Before preparing a response, confirm you understand the other person's actual concern, which is often not the first thing they said.

Sources & Further Reading

1. Orwell, G. (1946). Politics and the English Language. Horizon, 13(76).
2. Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257-285.
3. Minto, B. (1987). The Pyramid Principle: Logic in Writing and Thinking. Prentice Hall.
4. Heath, C., & Heath, D. (2007). Made to Stick: Why Some Ideas Survive and Others Die. Random House.
5. Tufte, E. R. (1983). The Visual Display of Quantitative Information. Graphics Press.
6. Tufte, E. R. (1997). Visual Explanations: Images and Quantities, Evidence and Narrative. Graphics Press.
7. Rosenberg, M. B. (2003). Nonviolent Communication: A Language of Life. PuddleDancer Press.
8. Nichols, M. P. (1995). The Lost Art of Listening. Guilford Press.
9. Pinker, S. (2014). The Sense of Style: The Thinking Person's Guide to Writing in the 21st Century. Viking.
10. Mayer, R. E. (2001). Multimedia Learning. Cambridge University Press.
11. Plain Language Action and Information Network. (2011). Federal Plain Language Guidelines. plainlanguage.gov.

Further Reading

• Meyer, E. (2014). The Culture Map. PublicAffairs.

Related reading: async communication explained, conflict communication explained, how to make better decisions