Review of van Fraassen pre-print


 

 

 
 

 

Edinburgh Research Explorer 
 
 

 
 

 
 

 
 

 

 
 

 
 

 
 

 
 

 
 

 
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Review of Bas van Fraassen.  Scientific representation:
paradoxes of perspective

Citation for published version:
Massimi, M 2009, 'Review of Bas van Fraassen. Scientific representation: paradoxes of perspective',
International Studies in the Philosophy of Science, vol. 23, no. 3, pp. 323-337.
https://doi.org/10.1080/02698590903197799

Digital Object Identifier (DOI):
10.1080/02698590903197799

Link:
Link to publication record in Edinburgh Research Explorer

Document Version:
Peer reviewed version

Published In:
International Studies in the Philosophy of Science

Publisher Rights Statement:
This is an Author's Accepted Manuscript of an article published in International Studies in the Philosophy of
Science (2009). The final version is available online at: http://dx.doi.org/10.1080/02698590903197799

General rights
Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s)
and / or other copyright owners and it is a condition of accessing these publications that users recognise and
abide by the legal requirements associated with these rights.

Take down policy
The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer
content complies with UK legislation. If you believe that the public display of this file breaches copyright please
contact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately and
investigate your claim.

Download date: 06. Apr. 2021

https://doi.org/10.1080/02698590903197799
https://doi.org/10.1080/02698590903197799
https://www.research.ed.ac.uk/portal/en/publications/review-of-bas-van-fraassen-scientific-representation-paradoxes-of-perspective(20ffb444-0883-4888-85df-10e0f771d31a).html


Scientific	
  Representation:	
  Paradoxes	
  of	
  Perspective	
  

Michela	
  Massimi	
  
University	
  College	
  London	
  
Published	
  in	
  International	
  Studies	
  in	
  the	
  Philosophy	
  of	
  Science	
  Vol.	
  23,	
  No.	
  3,	
  October	
  2009,	
  pp.	
  323–
337	
  
	
  
Scientific	
   Representation:	
   Paradoxes	
   of	
   Perspective;	
   Bas	
   C.	
   van	
   Frassen,	
   Oxford,	
   Clarendon	
   Press,	
  
2008,	
  xiv	
  +	
  303	
  pp.,	
  ISBN	
  9780199278220,	
  £30.00,	
  US$50.00	
  (hardback)	
  

Scientific	
  representation	
  is	
  a	
  fast-­‐growing	
  topic	
  in	
  contemporary	
  philosophy	
  of	
  science.	
  The	
  problem	
  
of	
  explaining	
  how	
  science	
  represents	
  the	
  world	
  is	
  an	
  old	
  one,	
  deeply	
  entangled	
  with	
  the	
  issue	
  of	
  
realism	
  and	
  the	
  problems	
  that	
  normally	
  come	
  with	
  it	
  (from	
  reference	
  to	
  mind	
  independence).	
  But	
  in	
  
recent	
  times,	
  the	
  topic	
  of	
  scientific	
  representation	
  has	
  taken	
  a	
  life	
  of	
  its	
  own,	
  mainly	
  because	
  of	
  the	
  
fashionable	
   experimentalist	
   quarters	
   from	
   which	
   it	
   originates.	
   Attention	
   to	
   scientific	
   practice	
   and	
  
how	
   scientific	
   models	
   represent	
   phenomena	
   raises	
   specific	
   questions	
   about	
   the	
   very	
   nature	
   of	
  
scientific	
  representation,	
  and	
  the	
  difference	
  between	
  representation	
  in	
  science	
  and	
  representation	
  
in	
  the	
  arts.	
  

Bas	
   van	
   Fraassen’s	
   book	
   is	
   the	
   most	
   up-­‐to-­‐date	
   and	
   paradigmatic	
   expression	
   of	
   this	
   new	
  
trend	
  in	
  philosophy	
  of	
  science,	
  for	
  at	
  least	
  two	
  main	
  reasons.	
  Firstly,	
  it	
  explicitly	
  acknowledges	
  the	
  
experimentalist	
  roots	
  of	
  the	
  problem,	
  as	
  it	
  arises	
  in	
  the	
  contemporary	
  literature.	
  Measuring	
  is	
  a	
  way	
  
of	
   representing,	
   and	
   scientific	
   representation	
   is	
   perspectival	
   in	
   the	
   same	
   sense	
   as	
   Dürer’s	
   ‘art	
   of	
  
measurement’:	
  drawing	
  in	
  perspective	
  is	
  itself	
  a	
  measurement	
  technique.	
  

Second,	
   and	
   for	
   the	
   first	
   time	
   as	
   far	
   as	
   I	
   am	
   aware,	
   van	
   Fraassen’s	
   book	
   clearly	
   marks	
   a	
  
distinction	
  between	
  the	
  problem	
  of	
  representation	
  as	
  it	
  appears	
  in	
  the	
  sciences,	
  and	
  as	
  it	
  appears	
  in	
  
philosophy.	
  In	
  particular,	
  it	
  marks	
  a	
  distinction	
  between	
  the	
  problem	
  as	
  it	
  originally	
  emerged	
  with	
  
the	
  scientific	
  revolution	
  at	
  the	
  time	
  of	
  Copernicus	
  and	
  Galileo;	
  and	
  the	
  subsequent	
  philosophical	
  re-­‐
elaboration	
  of	
  the	
  problem	
  with	
  Descartes,	
  whereby	
  ‘The	
  problem	
  initially	
  faced	
  in	
  the	
  sciences	
  was	
  
thus	
  transposed	
  into	
  one	
  pertaining	
  mind	
  and	
  matter’	
  (275).	
  And	
  while	
  the	
  philosophical	
  problem	
  of	
  
the	
  external	
  world—from	
  Descartes,	
  to	
  Kant,	
  to	
  Bradley—may	
  well	
  be	
  unsolvable,	
  philosophers	
  of	
  
science	
  should	
  confine	
  their	
  attention	
  to	
  the	
  specific	
  problem	
  of	
  explaining	
   ‘how	
  can	
  an	
  abstract	
  
mathematical	
  structure	
  represent	
  a	
  concrete	
  physical	
  entity?’	
  (243),	
  which	
  leads	
  van	
  Fraassen	
  back	
  
to	
  some	
  classical	
  discussions	
  about	
  empirical	
  adequacy	
  and	
  ‘saving	
  the	
  phenomena’,	
  enriched	
  with	
  
new	
  important	
  caveats	
  and	
  distinctions,	
  as	
  we	
  shall	
  see	
  below.	
  

The	
  whole	
  book	
  then,	
  as	
  I	
  see	
  it,	
  is	
  articulated	
  around	
  these	
  two	
  main	
  themes,	
  and	
  develops	
  
a	
  sophisticated,	
  intriguing,	
  subtle	
  line	
  of	
  argument	
  that	
  goes	
  from	
  the	
  experimentalist	
  roots	
  of	
  the	
  
problem	
  of	
  scientific	
  representation,	
  to	
  the	
  final	
  diagnosis	
  of	
  the	
  divide	
  between	
  philosophy	
  and	
  the	
  
sciences	
  on	
  this	
  specific	
  problem.	
  Hence	
  the	
  four	
  parts	
  into	
  which	
  the	
  book	
  is	
  divided,	
  and	
  of	
  which	
  I	
  
can	
  only	
  offer	
  a	
  very	
  brief	
  overview	
  below,	
  at	
  the	
  cost	
  of	
  leaving	
  out	
  many	
  details.	
  I	
  hope	
  at	
  least	
  
that	
  at	
  the	
  end	
  of	
  this	
  overview,	
  the	
  subtle	
  line	
  of	
  argument	
  that	
  goes	
  from	
  the	
  first	
  to	
  the	
  second	
  
aforementioned	
  theme	
  will	
  become	
  clear.	
  

The	
  first	
  part,	
  consisting	
  of	
  the	
  first	
  three	
  chapters,	
  is	
  on	
  representation	
  itself.	
  Going	
  back	
  to	
  
Nelson	
   Goodman	
   and	
   most	
   recent	
   debates	
   on	
   scientific	
   representation,	
   van	
   Fraassen	
   makes	
   the	
  
point	
   that	
   there	
   is	
   no	
   strong	
   argument	
   to	
   ban	
   resemblance	
   from	
   representation.	
   If	
   anything,	
  
representation	
   trades	
   on	
   selective	
   resemblances	
   for	
   their	
   usefulness,	
   where	
   ‘use’	
   or	
   ‘usefulness’	
  



encompasses	
  the	
   intention	
  of	
  the	
  creator,	
  coding	
  conventions	
   in	
  the	
  community,	
  and	
  the	
  way	
   in	
  
which	
  the	
  audience	
  takes	
  it,	
  among	
  other	
  things.	
  The	
  emphasis	
  is	
  on	
  the	
  pragmatics,	
  more	
  than	
  on	
  
the	
   syntax	
   or	
   semantics	
   of	
   representation:	
   ‘A	
   scientific,	
   technical,	
   or	
   artistic	
   representation	
   is	
   an	
  
artefact	
  …	
  something	
  constituted	
  as	
  a	
  cultural	
  object,	
  through	
  its	
  role	
  or	
  function,	
  bestowed	
  upon	
  it	
  
in	
  practice’	
  (30).	
  Not	
  only	
  is	
  representation	
  intentional,	
  in	
  the	
  sense	
  of	
  being	
  related	
  to	
  the	
  epistemic	
  
intentions	
   of	
   the	
   user.	
   It	
   is	
   also	
   perspectival,	
   in	
   the	
   same	
   sense	
   of	
   Alberti’s	
   and	
   Dürer’s	
   pictorial	
  
perspective:	
   representing	
   is	
   the	
   ‘art	
   of	
   measuring’,	
   i.e.	
   of	
   using	
   machines	
   and	
   engines	
   to	
   offer	
  
representations	
  of	
  phenomena	
  not	
  as	
  they	
  are,	
  but	
  as	
  they	
  appear	
  from	
  the	
  particular	
  vantage	
  point	
  
of	
  an	
  observer.	
  

The	
  second	
  part	
  of	
  the	
  book	
  (chapters	
  4–7)	
  expands	
  on	
  the	
  idea	
  of	
  representing	
  as	
  the	
  art	
  of	
  
measuring.	
   This	
   is	
   the	
   part	
   of	
   the	
   book	
   that	
   fleshes	
   out	
   the	
   experimentalist	
   roots	
   of	
   the	
  
contemporary	
   problem	
   of	
   scientific	
   representation.	
   The	
   emphasis	
   is	
   all	
   on	
   scientific	
   instruments,	
  
their	
  three	
  main	
  roles	
  (representative,	
  imitative,	
  and	
  productive)	
  and	
  two	
  main	
  ways	
  of	
  looking	
  at	
  
them:	
  either	
  as	
  engines	
  of	
  creation	
  or	
  as	
  windows	
  upon	
  the	
  invisible	
  world.	
  It	
  is	
  in	
  this	
  context	
  that	
  
we	
   encounter	
   some	
   of	
   van	
   Fraassen’s	
   familiar	
   discussions	
   about	
   electron	
   microscopes	
   and	
  
spectroscopes	
  creating	
  ‘new	
  phenomena,	
  truly	
  humanly	
  observable	
  phenomena’	
  (100),	
  as	
  opposed	
  
to	
  being	
  windows	
  into	
  the	
  unobservable	
  realm.	
  The	
  discussion	
  is	
  enriched	
  by	
  a	
  new	
  emphasis	
  on	
  
measurement,	
   both	
   from	
   a	
   historical	
   perspective	
   (going	
   back	
   to	
   Mach	
   on	
   thermometers,	
   and	
  
Poincaré	
  and	
  Einstein	
  on	
  time	
  and	
  length	
  measurements—chapter	
  5)	
  and	
  from	
  a	
  philosophical	
  one,	
  
whereby	
   aspects	
   of	
   the	
   notion	
   of	
   ‘measurement	
   outcome’—i.e.	
   intentionality,	
   indexicality,	
   and	
  
perspectivity—are	
  related	
  to	
  specific	
  kinds	
  of	
  representation	
  (especially	
  imaging	
  and	
  picturing,	
  180–
181).	
  

It	
  is	
  in	
  this	
  context	
  that	
  van	
  Fraassen	
  introduces	
  a	
  new	
  take	
  on	
  a	
  classical	
  distinction	
  of	
  his	
  
own,	
  namely	
  the	
  distinction	
  between	
  data	
  models	
  and	
  surface	
  models:	
  ‘the	
  data	
  model	
  summarizes	
  
the	
   relative	
   frequencies	
   found;	
   the	
   surface	
   model	
   “smoothes”—in	
   fact	
   “idealizes”—this	
   summary	
  
still	
  further	
  so	
  as	
  to	
  replace	
  the	
  relative	
  frequency	
  counts	
  by	
  measures	
  with	
  a	
  continuous	
  range	
  of	
  
values.	
  …	
  The	
  abstracting	
  is	
  an	
  idealising,	
  an	
  extrapolation	
  to	
  a	
  form	
  that	
  could	
  not	
  be	
  reached	
  in	
  
actual	
  practice’	
  (167	
  and	
  172).	
  This	
  distinction	
  turns	
  out	
  to	
  be	
  all	
  the	
  more	
  relevant	
  for	
  the	
  rest	
  of	
  
the	
  book,	
  in	
  particular	
  for	
  the	
  discussion	
  on	
  structuralism	
  that	
  occupies	
  Part	
  III.	
  

Indeed,	
  in	
  the	
  following	
  Part	
  III	
  of	
  the	
  book	
  (chapters	
  8–11),	
  the	
  focus	
  of	
  the	
  discussion	
  shifts	
  
from	
   measurement	
   as	
   representation	
   to	
   the	
   contention	
   that	
   scientific	
   representation	
   is	
   about	
  
structure	
   only.	
   Through	
   a	
   historical	
   excursus	
   via	
   Hertz’s	
   and	
   Boltzmann’s	
   Bildtheorie	
   to	
   Russell’s	
  
structuralism,	
   and	
   from	
   Carnap’s	
   Aufbau	
   to	
   Putnam’s	
   model-­‐theoretic	
   argument	
   against	
  
metaphysical	
  realism,	
  van	
  Fraassen	
  returns	
  to	
  another	
  familiar	
  topic	
  of	
  his	
  own,	
  namely	
  the	
  defence	
  
of	
  an	
  empiricist	
  version	
  of	
  structuralism.	
  Hertz’s	
  problem	
  of	
  explaining	
  ‘just	
  how	
  do	
  those	
  “pictures”,	
  
those	
  mathematical	
  constructions,	
  represent	
  what	
  they	
  represent?’	
  (208)	
  is	
  the	
  same	
  problem	
  that	
  
Newman	
   raised	
   against	
   Russell’s	
   structuralism;	
   and	
   the	
   very	
   same	
   problem	
   that	
   also	
   empiricist	
  
structuralism	
  ought	
  to	
  address,	
  whereby:	
  

Essential	
  to	
  an	
  empiricist	
  structuralism	
  is	
  the	
  following	
  core	
  construal	
  of	
  the	
  slogan	
  
that	
  all	
  we	
  know	
  is	
  structure:	
  	
  

I.	
  Science	
  represents	
  the	
  empirical	
  phenomena	
  as	
  embeddable	
   in	
  certain	
  abstract	
  
structures	
  (theoretical	
  models).	
  

II.	
  Those	
  abstract	
  structures	
  are	
  describable	
  only	
  up	
  to	
  structural	
  isomorphism.	
  …	
  	
  



Empiricist	
  structuralism	
  is	
  a	
  view	
  not	
  of	
  what	
  nature	
  is	
  like	
  but	
  of	
  what	
  science	
  is.	
  
(238)	
  

Van	
   Fraassen	
   uses	
   Putnam’s	
   model-­‐theoretic	
   argument	
   to	
   fight	
   the	
   metaphysical	
   realist	
   view	
  
(associated	
  with	
  the	
  correspondence	
  theory	
  of	
  truth)	
  claiming	
  that	
  ‘there	
  is	
  an	
  essentially	
  unique	
  
privileged	
   way	
   of	
   representing:	
   “carving	
   nature	
   at	
   the	
   joints”’	
   (244).	
   At	
   the	
   same	
   time,	
   he	
   uses	
  
Putnam’s	
  argument	
  also	
  against	
  non-­‐empiricist	
  versions	
  of	
  structuralism	
  that	
  tend	
  to	
  forget	
  that	
  ‘We	
  
have	
  an	
  interpretation	
  for	
  the	
  given	
  language	
  only	
  if	
  we	
  can	
  define	
  or	
  identify	
  such	
  a	
  function.	
  To	
  do	
  
that	
  we	
  must	
  be	
  able	
  to	
  describe	
  both	
  the	
  function’s	
  domain	
  and	
  its	
  range	
  …	
  .	
  As	
  long	
  as	
  we	
  are	
  not	
  
given	
  an	
  independent	
  description	
  of	
  both	
  the	
  domain	
  and	
  the	
  range	
  of	
  the	
  interpretation,	
  we	
  do	
  not	
  
have	
  any	
  such	
  interpretation,	
  nor	
  any	
  way	
  to	
  identify	
  one’	
  (233–234).	
  

But	
  while	
  Putnam,	
  led	
  by	
  the	
  model-­‐theoretic	
  argument,	
  landed	
  in	
  Kantian	
  internal	
  realism	
  
(until	
  the	
  pragmatic	
  turn	
  of	
  the	
  late	
  1980s),	
  van	
  Fraassen,	
  led	
  by	
  the	
  same	
  argument,	
  opts	
  for	
  a	
  self-­‐
declared	
   Wittgensteinian	
   move	
   with	
   an	
   emphasis	
   on	
   the	
  use	
  of	
   theories	
   and	
   representation	
   (see	
  
235),	
  in	
  continuity	
  with	
  the	
  pragmatics	
  of	
  scientific	
  representation	
  discussed	
  in	
  Part	
  I.	
  But	
  how	
  can	
  
such	
  a	
  Wittgensteinian	
  move	
  help	
  us	
  address	
  the	
  problem	
  of	
  explaining	
  ‘how,	
  or	
  in	
  what	
  sense,	
  can	
  
such	
   an	
   abstract	
   entity	
   as	
   a	
   model	
   “save”	
   or	
   fail	
   to	
   “save”	
   this	
   concrete	
   phenomenon?’	
   (245).	
  
Moreover,	
  if	
  ‘saving	
  the	
  phenomena’	
  implies	
  embedding	
  data	
  models	
  into	
  theoretical	
  models,	
  both	
  
of	
   which	
   are	
   abstract	
   structures,	
   then	
   ‘doesn’t	
   a	
   reflection	
   that	
   focuses	
   on	
   the	
   data	
   model	
   for	
  
assessing	
  empirical	
  adequacy,	
  lose	
  contact	
  with	
  reality	
  altogether?’	
  (246).	
  	
  

We	
  reach	
  here	
  what	
  in	
  my	
  view	
  is	
  the	
  most	
  interesting	
  part	
  of	
  the	
  book,	
  where	
  the	
  previous	
  
discussion	
   on	
   the	
   pragmatics	
   of	
   representation	
   and	
   on	
   measurement	
   as	
   representation	
   come	
  
together	
  and	
  some	
  classical	
  tenets	
  of	
  van	
  Fraassen’s	
  constructive	
  empiricism	
  are	
  revisited.	
  The	
  self-­‐
declared	
   ‘Wittgensteinian	
   move’	
   (254)	
   consists	
   then	
   in	
   the	
   following:	
   ‘the	
   theory	
   to	
   phenomena	
  
relation	
  displayed	
  here	
  is	
  an	
  embedding	
  of	
  one	
  mathematical	
  structure	
  in	
  another	
  one.	
  For	
  the	
  data	
  
model—or,	
   more	
   accurately,	
   the	
   surface	
   model—which	
   represents	
   the	
   appearances,	
   is	
   itself	
   a	
  
mathematical	
  structure	
  …	
  .	
  Construction	
  of	
  a	
  data	
  model	
  is	
  precisely	
  the	
  selective	
  relevant	
  depiction	
  
of	
   the	
   phenomena	
   by	
   the	
   user	
   of	
   the	
   theory	
  required	
   for	
   the	
   possibility	
   of	
   representation	
   of	
   the	
  
phenomenon	
  …	
  .	
  There	
   is	
  nothing	
   in	
  an	
  abstract	
  structure	
   itself	
  that	
  can	
  determine	
  that	
   it	
   is	
  the	
  
relevant	
  data	
  model,	
  to	
  be	
  matched	
  by	
  the	
  theory’	
  (252–253).	
  

By	
  shifting	
  attention	
  to	
  the	
  user-­‐dependent	
  notion	
  of	
  scientific	
  representation,	
  van	
  Fraassen	
  
can	
  maintain:	
  

For	
  us	
  the	
  claims	
  

(A)	
  that	
  the	
  theory	
  is	
  adequate	
  to	
  the	
  phenomena	
  and	
  the	
  claim	
  

(B)	
  that	
  it	
  is	
  adequate	
  to	
  the	
  phenomena	
  as	
  represented,	
  i.e.	
  as	
  represented	
  by	
  us	
  

	
  are	
  indeed	
  the	
  same!	
  That	
  (A)	
  and	
  (B)	
  are	
  the	
  same	
  for	
  us	
  is	
  a	
  pragmatic	
  tautology.	
  
(259)	
  

And	
  again:	
  ‘in	
  a	
  context	
  in	
  which	
  a	
  given	
  model	
  is	
  someone’s	
  representation	
  of	
  a	
  phenomenon,	
  there	
  
is	
  for	
  that	
  person	
  no	
  difference	
  between	
  the	
  question	
  whether	
  the	
  theory	
  fits	
  that	
  representation	
  
and	
  the	
  question	
  whether	
  that	
  theory	
  fits	
  the	
  phenomena’	
  (260).	
  

Having	
  taken	
  this	
  crucial	
  pragmatic	
  step,	
  in	
  the	
  final	
  Part	
  IV,	
  van	
  Fraassen	
  finally	
  returns	
  to	
  
the	
  issue	
  of	
  realism.	
  As	
  anticipated	
  above,	
  in	
  this	
  final	
  part	
  of	
  the	
  book	
  he	
  explores	
  the	
  parting	
  of	
  the	
  



way	
  in	
  the	
  problem	
  of	
  representation	
  as	
  it	
  emerged	
  in	
  the	
  sciences	
  with	
  Galileo,	
  and	
  as	
  it	
  was	
  later	
  
reinterpreted	
   by	
   Descartes	
   and	
   other	
   philosophers	
   as	
   the	
   ‘problem	
   of	
   the	
   external	
   world’.	
   By	
  
building	
  up	
  on	
  the	
  previous	
  user-­‐dependent,	
  perspectival	
  analysis	
  of	
  scientific	
  representation,	
  van	
  
Fraassen	
  now	
  introduces	
  a	
  new	
  distinction	
  between	
  appearances	
  and	
  phenomena:	
  ‘Phenomena	
  are	
  
observable,	
   but	
   their	
   appearance,	
   that	
   is	
   to	
   say,	
   what	
   they	
   look	
   like	
   in	
   given	
   measurement	
   or	
  
observation	
  set-­‐ups	
  is	
  to	
  be	
  distinguished	
  from	
  them	
  as	
  much	
  as	
  any	
  person’s	
  appearance	
  is	
  to	
  be	
  
distinguished	
   from	
   that	
   person’	
   (284).	
   Appearances	
   are	
   simply	
   the	
   contents	
   of	
   measurement	
  
outcomes,	
   and	
   should	
   not	
   be	
   confused	
   with	
   what	
   philosophers	
   sometimes	
   calls	
   ‘appearances’,	
  
namely	
  subjectively	
  experienced	
  impressions	
  (see	
  276).	
  

On	
  the	
  other	
  hand,	
  phenomena	
  are	
  observable	
  things	
  and	
  events	
  in	
  the	
  world:	
  they	
  are	
  the	
  
‘smelly,	
  colourful,	
  noisy	
  things’	
  which	
  are	
  real	
  (276).	
  A	
  paradigmatic	
  example	
  is	
  how	
  Copernicus’s	
  
theory	
   saved	
   the	
   ‘phenomenon’	
   of	
   Mercury’s	
   motion,	
   by	
   showing	
   how	
   the	
   ‘appearance’	
   of	
   its	
  
retrograde	
  motion	
  could	
  be	
  derived	
  (via	
  kinematics	
  and	
  optics)	
  from	
  what	
  Copernicus	
  postulated	
  
about	
  Mercury’s	
  motion.	
  One	
  may	
  also	
  be	
  tempted	
  to	
  equate	
  van	
  Fraassen’s	
  distinction	
  between	
  
appearances	
  and	
  phenomena	
  to	
  the	
  previous	
  distinction	
  between	
  data	
  models	
  and	
  surface	
  models,	
  
although	
  van	
  Fraassen	
  does	
  not	
  do	
  this	
  explicitly.	
  The	
  bottom	
  line	
  is	
  that	
  reality	
  consists	
  of	
  smelly,	
  
colourful,	
  noisy	
  (observable)	
  phenomena	
  (not	
  Kantian	
  things-­‐in-­‐themselves,	
  nor	
  a	
  Cartesian	
  external	
  
world),	
  while	
  appearances	
  are	
  the	
  way	
  phenomena	
  ‘look	
  like’	
  in	
  a	
  given	
  measurement	
  set-­‐up,	
  and	
  
hence	
  from	
  a	
  particular	
  vantage	
  point	
  (as	
  in	
  Dürer’s	
  pictorial	
  perspective).	
  We	
  ‘save	
  phenomena’	
  by	
  
embedding	
   perspectival	
   appearances	
   (as	
   given	
   by	
   a	
   certain	
   instrument,	
   measurement	
   set-­‐up,	
   or	
  
frequencies	
  in	
  a	
  data	
  model)	
  into	
  another	
  abstract	
  structure,	
  the	
  surface	
  model,	
  which	
  ‘smoothes’	
  
and	
   ‘idealises’	
   the	
   measurement	
   outcomes,	
   and	
   eventually	
   embed	
   the	
   surface	
   model	
   into	
  
theoretical	
  models,	
  such	
  as	
  for	
  example	
  Copernicus’s	
  geometric	
  models	
  in	
  astronomy.	
  

This	
  brief	
  overview	
  of	
  the	
  book	
  does	
  not	
  do	
  justice	
  to	
  the	
  complexity,	
  details,	
  and	
  nuances	
  of	
  
it.	
   But	
   I	
   hope	
   it	
   suffices	
   to	
   give	
   an	
   idea	
   of	
   the	
   fascinating	
   journey	
   that	
   goes	
   from	
   a	
   serious	
   re-­‐
appraisal	
  of	
  the	
  role	
  of	
  scientific	
   instruments	
  in	
  artistic	
  and	
  scientific	
  representation,	
  to	
  the	
  more	
  
general	
  issue	
  of	
  how	
  science	
  represents	
  nature,	
  and	
  how	
  its	
  modalities	
  and	
  operations	
  should	
  not	
  be	
  
conflated	
  with	
  the	
  philosopher’s	
  problem	
  of	
  the	
  external	
  world,	
  according	
  to	
  van	
  Fraassen.	
  This	
  is	
  a	
  
book	
  with	
  an	
  important	
  and	
  intricate	
  story	
  to	
  tell,	
  and	
  it	
  does	
  it	
  with	
  so	
  much	
  attention	
  to	
  historical,	
  
philosophical,	
   scientific,	
   and	
   even	
   artistic	
   details	
   that	
   makes	
   it	
   an	
   occasion	
   for	
   a	
   thousand	
   wider	
  
meditations.