Transcript Document

Clio Meets Seshat:
Building the Global History Database
Peter Turchin
Dublin, June 2014
A Science of History?
• According to most historians, history is
a part of the humanities
• Most historians have abandoned the
belief in general laws
• Yet, when historians construct
narratives they also propose
explanations for why things happened
the way they did
– which implies existence of general
principles (“laws”)
History as viewed by
a natural scientist
• A mature descriptive discipline that
requires high technical expertise
• But it is not (yet) a theoretical,
explanatory science
• History needs
– a falsificationist agenda
– a mathematical component
– systematic databases for testing models
Why History Needs Mathematics
• A science becomes Science only after it
gains mathematical content
– formal models
– statistical analysis
• Why: to translate assumptions into
predictions (for empirical testing)
– especially in nonlinear dynamics
• Explicit mathematical models can correct
faulty verbal theory
– example: the theory of “imperial overstrech”
• An empire gobbles up
too much territory,
incurs heavy
logistical burdens
that cause it to
collapse
– Paul Kennedy
– Randall Collins
X(t)
Imperial Overstrech: the Theory
t
dA
 cA exp[ A / h ]  a
dt
Logistical
loads
L
+
Territory
size
A
Geopolitical
resources
R
+
Conclusion: theory of
imperial overstretch
leads to a first-order
differential equation
that cannot exhibit
boom-bust dynamics
+
War
success
W
X(t)
+
predicted dynamics
-
t
Why do
Empires Fall?
“The Decline
and Fall of the
Roman Empire”
“My name is Ozymandias,
king of kings...”
Why did the Roman Empire Fall?
• The German historian Alexander
Demandt counted at least 210
explanations of why Rome fell
• Demandt, A. 1984. Der Fall Roms: die Auflösung
des Römischen Reiches im Urteil der Nachwelt
(Beck, Munich)
• The problem with history, as it is
traditionally practiced, is that theories
multiply but are never rejected
… and
explanations
continue to
multiply…
Can ancient
history tell us
anything about
today?
Why we need to start
reject hypotheses
• In natural sciences progress occurs
when some hypotheses/theories are
rejected in favor of others
– Phlogiston
– Lamarkism
The Good Old Scientific Method
• Define the question
• Propose two or more alternative
explanations/theories
• Use mathematical models to extract
predictions from theories
– predictions that disagree about some
observable aspect of reality
• Put together data to adjudicate
between the theories
• Repeat as necessary
The Puzzle of Ultrasociality
• Ultrasociality –
extensive
cooperation among
very large numbers
of genetically
unrelated
individuals
• How did it evolve?
International Space Station
Approaches:
• General theory: cultural multilevel
selection (CMLS) of ultrasocial norms
and institutions
• A specific model: Africa and Eurasia,
1500 BCE – 1500 CE
• Empirical tests: building a massive
historical database of cultural evolution
General Theory
definitions
• Ultrasociality: extensive cooperation among
very large numbers (e.g. >106) of genetically
unrelated individuals
• Norms: culturally acquired rules of behavior
• Institutions: systems of norms that govern
behavior of individuals in specific contexts
• Ultrasocial norms and institutions: provide
the basis for integration of large-scale
societies, but have costs for lower-level
units
Examples of ultrasocial norms
• Propensity to trust and help individuals
outside one’s ethnic group (“generalized
trust”)
– benefit: provides a basis for cooperation in
multiethnic societies
– cost: vulnerability to free-riding by ethnic groups
that restrict cooperation to coethnics
•
•
•
•
Willingness to pay national taxes
Obeying laws
Refusing bribes and not offering bribes
Volunteering for military service in times of
war
Examples of ultrasocial institutions
• Government by professional bureaucracies
– basis for one common definition of the state
– benefit: governing sufficiently large-scale
societies is apparently impossible without
bureaucrats, record-keeping, division of tasks
– cost: expensive to train and maintain bureaucrats;
principal-agent problems
• Universal religions and other integrative
ideologies
• Legitimating power/restraining rulers
• The state as a ‘bundle’ of ultrasocial
institutions
Understanding how ultrasocial traits spread
• is not a simple matter of accounting for their
benefits for integration of large-scale
societies
• these institutions have significant costs
– and historical record indicates that they
repeatedly collapsed
• need an evolutionary mechanism to explain the
spread of such traits despite the costs
• CMLS: cultural multilevel selection
– “group selection”
– Boyd, Richerson, D.S. Wilson, Bowles, Turchin
Major
Evolutionary
Transitions:
•
•
•
•
Eukaryotic cell
Multicellular organism
Eusocial insect colony
Complex human society
–
–
–
–
–
• General Processes
“particle” cooperation
selection on “collectives”
suppression of particle selfishness
and competition
increasing functional integration
of collectives
collectives become organisms
A Social Scale (Agrarian Polities)
Population
10,000,000s
1,000,000s
100,000s
10,000s
1,000s
100s
Area, km2
Polities
1,000,000s Mega-empires
100,000s Macrostates
10,000s States (Archaic),
Supercomplex chiefdoms
1,000s Complex chiefdoms,
City states
100s Simple chiefdoms,
acephalic tribes
Local communties (villages)
Alternative Theories
• Resource base (agriculture)
– Childe, White, Service, Diamond
• Social differentiation and class structure
– Marx, Engels, Patterson
• Warfare and circumscription - Carneiro
• Cultural Multilevel Selection
– Boyd, Richerson, D.S. Wilson, Bowles
• Economics and trade, problem-solving and
information processing, …
Real
Data
Overall model fit
R2 ≈ 0.65
Simulated
Data
Spread of ultrasocial traits predicted by the model
SESHAT: Global History Databank
The huge corpus of knowledge about past societies collectively
possessed by academic historians is almost entirely in a form that is
inaccessible to scientific analysis, stored in historians’ brains or scattered
over heterogeneous notes and publications. The huge potential of this
knowledge for testing theories about political and economic
development has been largely untapped.
Our goal: a historical database that will enable us and others to test
theories about the processes responsible for the rise of large-scale
societies in human history. The database will bring together, in a
systematic form, what is currently known about the sociopolitical
organization of human societies, and how it has evolved with time.
An example: bureacracy
characteristics
• Examination system
• Merit promotion
• Solutions to the
principal-agent problem
SESHAT: Global History Databank
Editorial Board
Peter Turchin (UConn): overall coordinator; social complexity
Harvey Whitehouse (Oxford): co-editor; ritual and religion
Pieter François (Oxford): historical coordinator; ritual variables
Thomas Currie (Exeter): resources, agriculture, and population
Kevin Feeney (TCD): information technology
Consultants
J. G. Manning (Yale)
Douglas White (UC Irvine)
Arkadiusz Marciniak (Poznan)
Peter Peregrine (Lawrence and Santa Fe Institute)
Enrico Spolaore (Tufts)
David Sloan Wilson (Binghamton)
Peter Richerson (UC Davis)
Postdocs
Daniel Hoyer, and 2 postdocs to be hired
Research Assistants
Rudolf Cesaretti, Edward Turner, and ~10 short-term RAs
What will Seshat (eventually) do?
Feedback
databases
Electronic Archives
Collective
intelligence
Community of experts &
volunteers
Seshat
Databank
High
Quality
Open
Data
Data
Consumers
“improve the extraction of collective
intelligence from electronic archives,
research communities and data consumers
to improve the quality of published data”
SESHAT: Global History Databank
Acknowledgments
Bernard Winograd
Jim Bennett
Tricoastal Foundation