Leadership is Language - Notes

David Marquet is a top graduate of the US Naval Academy, he commanded the nuclear powered fast attack submarine Santa Fe from 1999 – 2001. A command which he documented in his best selling first book, Turn the Ship Around.

Worked through most of his navy service command with a prove and perform leadership style, make decisions get the team to implement them but making all the decisions absolves the team from thinking.

Had to change his style because taking command of Sante Fe was a last minute change and he did not have the normal 1-year lead in to learn the boat. He therefore did not know the boat and to use the crew in a different and ultimately better way, for everyone.

He noticed his language changed in 3-ways:

Moved from convince, coerce comply language to intent and commitment to action

Moved from prove and perform to improve and learn

Moved from invulnerability and certainty to vulnerability and curiosity.


It takes a lot of self-awareness and re-programming to move away from this default imperative mode of communication.


The book is structured around a model of America football plays and the idea that our current plays need updating.  The 6-new plays in the book are:

Control the clock instead of obeying the clock

Collaborate instead of coerce

Commitment rather than compliance

Complete defined goals instead of continuing work indefinitely

Improve outcomes rather than prove ability

Connect with people instead of conforming to your role.

Book takes the example of a container ship El Farro that sank with all hands when she sailed into a hurricane in the Caribbean. It’s a great case study as the bridge voice transcripts were recovered which illustrate many of the points about how our language pre-determines bad choices and actions. https://en.wikipedia.org/wiki/SS_El_Faro

Use a share of voice analysis to note the relative imbalance / hierarchy gradient amongst the crew vs Capt.

Compares the old way of doing things to a 20th century Winslow Taylor inspired approach which seeks to reduce variability  – great in a factory and a 21st century thinking approach which seeks to increase variability in the thinking process. This sort of language is more open, curious and probabilistic.

Marquet describes these approaches as red work (doing) and blue work (thinking) his point is that to be successful we need to continually alternate between these different phases.  However our leadership language tends to be anachronistic and much more focused on red work, its about doing not thinking and it tends to reduce variability. Get it done, make it happen, lets finish this etc. In the 20th century different types of people did these separate tasks, workers did the doing and managers did the thinking.

Thinking language sounds more like, how do you see this? How ready are we for this? What can we do better? What did we learn?

W Edwards Deming worked with the Japanese to rebuild their economy post WW2 especially in the automobile industry – he developed the total quality management approach. Quality of products came from reducing variability but Deeming’s system built QA into the manufacturing system not as previously leaving it to the end when parts might get rejected. This involved lots of experimentation up front to optimize the system for quality but the best people to help drive this are not managers but the assembly line staff themselves.

In the airline industry Crew resource management (CRM) is another response to help improve safety though how flight deck staff interact and communicates to allow concerns to be voiced and to counter the authority bias of the Captains position.

Doing work is relatively immune to stress – in fact applying stress is a common management motivational technique in this space. Thinking work is however highly susceptible to stress because it impairs the pre-frontal cortex which is drained of blood when the brain initiates a threat response. A protect mindset where pro-social behaviours are often diminished.

Controlling the clock

Makes pauses for thinking and reviewing possible. Give the pause named, time out, hands off, yellow card etc. Also needs practice to acclimatize people that this is fine.

Toyota use the Andon cord a Japanese word for a traditional paper lantern. It allows any worker to call attention to a problem.

When the engineers from Morton-Thiokol the space shuttle booster manufacturers recommended another delay in launch due to cold weather , NASA officials said they were appalled, ‘My God, when do you want me to launch, April.’ – Exaggeration, hyperbole and exasperation all about obeying the clock.

We have time to do this right not twice.

Controlling the clock allows us to phase out of red work and into blue so it starts the cycle.

Moving from coercion to collaboration.

Vote first, then discuss

Be curious, not compelling

Invite dissent rather than drive consensus

Give information not instructions.

Typical compliance language

Because I was told to

Because it says so

I just do what they tell me

I’m not paid to think

From Compliance to Commitment

Commit to learn not just do

Commit actions not beliefs

Chunk it small but do it all

Commitment is better than compliance because it releases discretionary effort.

Moving from Continue to complete

Chunk work for frequent completes early and few completes late

Celebrate with not far

Focus on behaviour not characteristics

Focus on journey, not destination

Continuous improvement is a misnomer focus instead on discontinuous improvement.

Motivational factors – Competence, relatedness and autonomy.(Self determination theory, Deci & Ryan)

Overcoming learned helplessness

Frequent complaint of leaders is that they are giving out control but that people don’t want it. Could be lots of reasons for this but learned helplessness may be a part of it – the idea that nothing you do matters.

Language can help to challenge this to avoid a view of negative events that sees them as permanent , personal and pervasive – this would always happen to me.

Looking forward.

What do we want to do differently next time?

If we could go back in time what would we tell ourselves?

What do we want to remember for next time?

Focus outward – how to help others

If someone else was going to take over this project what would you say to them to help?

What could we do to better serve our customers?

Continuous improvement

Let’s focus on what is going right here so we can build on that.

Improve as a play is about reflecting on what you’ve done and how to make it even better. Improve is about effective collaboration and links into the connect play.



Connect is about caring. Caring what people think and feel and caring for their personal goals. This play does not fit into the cycle of red and blue work but is rather a background building block.  Key factor in the environment is the nature of the power gradient.

To do this

Flatten the power gradient

Admit you don’t always know

Be vulnerable

Trust first.

Censoring information will be proportionate to the power gradient. Where steep power gradients exisit employees will carefully censor information upwards. (Good diagnostic question, when was the last time a lower ranked employee tell you something you really did not want to hear?)

Share of voice is also very indicative of the power gradient.

Too often we wont trust until the other party ‘proves’ themselves trustworthy. But his sets us up in a mode of judgement instead of observation, increasing the power gradient. We should assume good intent.

Strict goals + steep hierarchies = unethical behaviour.

Book closes with the type of conversations that could have happened on the Bridge and saved the El Farro. This time driving the right level of reflection, collaboration and informed decision-making.