001 / From the record
Every idea a lens uses is tied to a work and the year it appeared, with links to the text or its publisher.
The lenses
Ten lenses. Each asks one question.
A lens is one way of reasoning, named for what it tests and built from one person’s published work. A review uses three, picked for the argument in front of it. None reasons from anything later than 1980.
Lenses · in six rows
001Does the conclusion follow?
001 / RefutationWhat observation would prove this wrong?
What conceivable observation would count against this claim?
From the work of Karl Popper · reasons from 1934–1980002 / Formal modelDo the model’s assumptions match the thing modelled?What are the incentives, available strategies and consequences of each move?
From the work of John von Neumann · reasons from 1928–1957002Is the mechanism real?
003 / MechanismWhat experiment separates the explanation from a good story?
What experiment would distinguish this explanation from a convincing story?
From the work of Richard Feynman · reasons from 1948–1980004 / The exceptionWhat does the outlier show that the average hides?What does the exception reveal that the average has hidden?
From the work of Barbara McClintock · reasons from 1931–1980005 / FeedbackDoes measuring or acting on the system change it?How does the output change the system that produced it?
From the work of Norbert Wiener · reasons from 1943–1964003Is the measurement sound?
006 / Signal and noiseWhich distinctions survive the measurement, and which are lost?
Which distinctions survive transmission, and which are lost?
From the work of Claude Shannon · reasons from 1937–1961007 / Operational testWhat would a passed test actually establish?What observable test would let us distinguish the competing claims?
From the work of Alan Turing · reasons from 1936–1954004Does it hold elsewhere?
008 / Sampling frameWho or what was left out of the sample?
What becomes visible from the position the dominant account excludes?
From the work of W. E. B. Du Bois · reasons from 1897–1963009 / Bounded searchWhat did the real decision process have time to consider?How does a person decide with limited time, attention and information?
From the work of Herbert Simon · reasons from 1947–1980010 / UncertaintyWhich unknowns can the stated odds not cover?Which future possibilities cannot honestly be reduced to known probabilities?
From the work of John Maynard Keynes · reasons from 1919–1946005How a lens is built.
002 / Held to a horizon
A lens reasons only with ideas documented inside its horizon. None runs past 1980, before personal computers and the web.
003 / Never the person
A lens never speaks as them or quotes them, and each names a documented blind spot, so you know where not to trust it.
006Choose them yourself.
Feynman picks three lenses for each idea, to suit the argument in front of it. A member can change the choice before the review starts.