1610.00132

## Charles  Sanders  Peirce  and  the  Abduction  of Einstein: On  the  Comprehensibility  of  the  World 1 Cornelis  de  Waal cdwaal@iupui.edu ## Abstract Einstein  was  deeply  puzzled  by  the  success  of  natural  science,  and  thought  that  we  would never  be  able  to  explain  it.  He  came  to  this  conclusion  on  the  ground  that  we  cannot  extract the  basic  laws  of  physics  from  experience  using  induction  or  deduction,  and  he  took  this  to mean  that  they  cannot  be  arrived  at  in  a  logical  manner  at  all.  In  this  paper  I  use  Charles Peirce's  logic  of  abduction,  a  third  mode  of  reasoning  different  from  deduction  and induction,  and  show  that  it  can  be  used  to  explain  how  laws  in  physics  are  arrived  at, thereby  addressing  Einstein's  puzzle  about  the  incomprehensible  comprehensibility  of  the universe.  Interpreting  Einstein's  reflections  in  terms  of  Peirce's  abduction  also  sheds  light on  abduction  itself,  by  seeing  it  applied  in  an  area  where  our  common  sense,  and  with  that our  intuitions,  give  us  little  or  no  guidance,  and  is  even  prone  to  lead  us  astray. Einstein  was  deeply  puzzled  by  the  success  of  natural  science  and  thought  that  we would  never  be  able  to  explain  it.  'The  eternally  incomprehensible  thing  about  the world,'  he  famously  quipped,  'is  its  comprehensibility.' 2  Unable  to  account  for  this, Einstein  dubbed  it  a  miracle  ( Wunder )  and  remarked  that  it  instilled  in  him  a 'cosmic  religious  feeling.' 3  Moreover,  he  added,  the  more  we  learn  about  the universe,  the  more  incomprehensible  it  becomes  that  we  can  draw  the  conclusions 1  An  earlier  version  of  this  paper  was  read  at  the Congress. Charles  S.  Peirce  2014  Centennial Lowell,  Mass.,  July  16-19,  2014.  I  would  like  to  thank  the  members  of  the audience  for  their  valuable  criticisms  and  suggestions. 2  Albert  Einstein, Ideas  and  Opinions Ideas  and  Opinions (New  York:  Bonanza  Books,  1954),  292. ,  39. we  do. 4  Einstein's  notion  of  comprehension  is  a  modest  one;  it  is  merely  'the production  of  some  sort  of  order  among  sense  impressions.' 5 In  part,  Einstein's  view  that  the  comprehensibility  of  the  world  is  an  'eternal secret,' 6  is  the  result  of  his  take  on  how  natural  science,  especially  physics,  works.  In this  presentation  I  will  address  Einstein's  puzzlement  by  utilizing  Charles  Peirce's logic  of  abduction.  My  argument  runs  roughly  as  follows.  First,  I  discuss  Einstein's views  on  the  aim  of  physics  and  his  resultant  criticisms  of  phenomenological  and empiricist  approaches  to  physics.  A  consequence  of  this  approach  is  that  basic physical  laws  cannot  be  extracted  from  experience  using  induction  or  deduction, and  for  Einstein  this  means  that  they  cannot  be  arrived  at  in  a  logical  manner  at  all. Peirce,  as  is  well  known,  also  distinguished  a  third  mode  of  reasoning,  which  he termed  abduction  or  retroduction.  This  raises  the  question  whether  this  third  mode of  reasoning  can  be  used  to  adequately  capture  how  in  Einstein's  view  we  arrive  at these  basic  laws.  If  so,  then  we  can  establish  that  how  these  laws  are  being  arrived at  is  a  logical  process  after  all,  and  Peirce's  justification  of  abduction  as  a  mode  of reasoning  can  be  used  to  address  Einstein's  puzzle  as  to  the  incomprehensible comprehensibility  of  the  world.  Interpreting  Einstein's  reflections  in  terms  of Peirce's  abduction  also  sheds  light  on  abduction  itself  by  seeing  it  applied  in  an  area where  our  common  sense,  and  with  that  our  intuitions,  give  us  little  or  no  guidance and  is  even  prone  to  lead  us  astray. ## 1.  The  Aim  and  Method  of  Physics For  Einstein,  the  aim  of  science  is  'to  make  the  chaotic  diversity  of  our  sense-­experience  correspond  to  a  logically  uniform  system  of  thought.' 7  In  doing  this,  the ideal  is  cover  as  much  of  our  sense  experiences  as  possible,  while  reducing  the number  of  primary  concepts  and  primary  relations  to  a  minimum.  In  line  with  this, Einstein  writes  in  his  1933  Herbert  Spencer  lecture  that  'a  complete  system  of 4  30  March  1952  letter  to  Maurice  Solovine;  reproduced  in  Albert  Einstein, Solovine (New  York:  Philosophical  Library,  1987). Ideas  and  Opinions 6  Ibid. Ideas  and  Opinions ,  292. ,  323. Letters  to theoretical  physics  is  made  up  of  [three  things:]  concepts,  fundamental  laws  which are  supposed  to  be  valid  for  those  concepts,  and  conclusions  to  be  reached  by  logical deduction.' 8 This  way  of  looking  at  science  is  reminiscent  of  the  view  Heinrich  Hertz developed  earlier  in  his Principles  of  Mechanics . 9  Hertz's  description  of  classical mechanics  may  serve  as  an  example.  Hertz  listed  four  primary  concepts  (space, time,  force,  and  mass)  and  four  primary  relations,  or  fundamental  laws,  to  connect them  (Newton's  three  laws  of  motion  and  d'Alembert's  principle).  Once  these primary  concepts  and  relations  are  given,  Hertz  argued,  everything  else  is  just  a matter  of  deductive  inference. 10  What  makes  those  concepts  and  relations  primary is  that  they  cannot  be  justified  by  the  theory  in  which  they  feature;  from  the perspective  of  the  theory  they  are  purely  given  and  non-­-justifiable.  Hertz  next presented  three  criteria  for  evaluating  such  axiomatic  systems  of  primary  concepts and  relations:  consistency,  correctness,  and  appropriateness.  The  axiomatic  systems should  be  logically  permissible,  not  contradict  observations,  and  in  cases  where multiple  theories  meet  both  demands,  represent  more  of  the  relations  we  deem important  and  less  relations  that  lack  any  counterpart  in  our  observations. With  regard  to  theoretic  structures  thus  conceived,  Einstein  argued  that  one should  opt  for  theories  with  primary  principles  and  concepts  that  are  simple,  few  in number,  and  where  the  coordination  with  sense  experience  is  both  unique  and convincing. 11  Moreover,  the  coordination  should  be  with  sense  impressions in  their totality rather  than  be  restricted  to  a  certain  predetermined  subset.  Put  differently, Einstein  is  looking  for  a  single,  global  theory,  not  for  a  bunch  of  separate  theories that  are  logically  independent  of  one  another. 12  A  comparison  of  Kepler's  laws  with Newton's  theory  of  gravitation  can  function  as  an  example.  Kepler's  three  laws  (that the  orbit  of  a  planet  is  an  ellipse  with  the  Sun  at  one  of  its  foci;  that  a  line  joining  a planet  and  the  sun  sweeps  out  equal  areas  during  equal  intervals  of  time;  and  that Ideas  and  Opinions 9  Heinrich  Hertz, ,  272. Principles  of  Mechanics 10  Op.  cit.,  4-5. Ideas  and  Opinions ,  323. Ideas  and  Opinions ,  293. (London:  Macmillan,  1899),  see  esp.  1-4. the  square  of  the  orbital  period  of  a  planet  is  proportional  to  the  cube  of  its  mean distance  to  the  sun)  are  three  brilliant  empirical  generalizations  that  paint  a  pretty complete  picture  of  the  motions  of  the  planets,  but  they  are  logically  independent. What  Newton  brings  to  the  table,  in  Einstein's  view,  is  one  single  theory  that  brings the  three  laws  together  by  proving  them  theorems  of  that  theory,  as  here  all physical  events,  including  the  movements  of  the  planets,  are  put  in  terms  of  masses that  are  subject  to  Newton's  laws  of  motion. 13  As  Einstein  observes,  this  move toward  greater  logical  unification  comes  at  a  cost.  The  more  comprehensive  the framework  becomes,  the  further  its  primary  concepts  and  relations  are  removed from  the  sense  experience  that  inspired  them  and  forms  their  touchstone.  This  is clearly  the  case  here,  where  we  move  from  Kepler's  geometrical  description  of  the motion  of  the  planet  Mars,  which  applies  to  the  other  planets  also,  to  the  highly abstract  ideas  of  force  and  mass  that  are  defined  contextually  in  terms  of  the theory's  other  primary  elements.  Einstein's  own  theory  of  relativity  takes  a  step further  in  this  direction. 14  In  the  end,  Einstein  writes,  we  get  'a  system  of  the greatest  conceivable  unity,  and  of  the  greatest  poverty  of  concepts  of  the  logical foundations.' 15  Hence,  rather  than  an  inference  to  a  plausible  hypothesis  (as  Peirce does)  or  an  inference  to  the  best  explanation  (as  Harman  and  Lipton  do),  the  search for  primary  concepts  and  principles  is  best  described  as  an  inference  to  the  greatest comprehension. ## 2.  The  Quest  for  Primary  Concepts  and  Primary  Principles In  his  discussion  of  mechanics,  Hertz  said  precariously  little  about  how  we  find these  primary  concepts  and  principles  from  which  all  else  is  supposed  to  follow deductively. 16  The  task  at  hand  is  to  specify  a  small  number  of  logically  independent Ideas  and  Opinions ,  257. Ideas  and  Opinions ,  349. Ideas  and  Opinions ,  294. 16  Admittedly,  Hertz  was  primarily  concerned  with  the  question  what  justifies  us  in maintaining  certain  systems,  and  not  in  the  question  how  its  primary  concepts  and relations  were  discovered.  As  Hertz  well  knew,  how  the  primary  concepts  and relations  were  discovered  is  immaterial  to  the  system's  justification. conceptual  elements  in  terms  of  which  all  our  sense  experience  can  be  ordered;  that is,  to  find  elements  that  are  simple  and  can  be  considered  basic.  For  instance,  in Newtonian  mechanics  velocity  is  not  a  logically  independent  concept,  as  it  is  a function  of  space  and  time.  Now  how  do  we  find  such  basic  elements?  One possibility  is  that  we  extract  them  from  experience.  This  is  in  essence  the empiricist's  answer.  Einstein  ascribes  this  view  to  Newton  whom  in  Einstein's  view 'still  believed  that  the  basic  concepts  and  laws  of  his  system  could  be  …  deduced from  experience  by  'abstraction'-that  is  to  say,  by  logical  means.' 17  For  instance, we  would  derive  the  primary  concept  of  space  a  posteriori  by  mentally  removing  all content  from  our  experience,  thus  creating  the  conception  of  empty  space.  By making  it  a logical derivation,  the  empiricist  further  implies  that  the  concepts  and principles  arrived  at  are  uniquely  determined  ( eindeutig  Bestimmt )  by  our perception  of  the  external  world. Einstein  rejected  the  first  answer.  In  fact,  he  rejected  that  the  process  is  a  logical one  at  all,  whether  deductive  or  inductive,  and  he  denied  that  the  result  is  uniquely determined.  Observing  that  abstraction  and  generalization  require  us  to  make choices,  the  process  cannot  be  a  logical  one  and,  moreover,  allows  for  alternatives. Consequently,  the  most  we  can  say  is  that  our  sense  impressions  inspire,  or  hint  at, certain  concepts  and  principles.  In  Einstein's  view,  this  element  of  choice  suffices  to make  the  resulting  conceptions  free  creations  of  the  human  mind. 18  To  this  we  can further  add,  as  was  said  before,  that  when  our  theories  become  more comprehensive,  like  with  the  shift  from  Kepler  to  Newton,  the  primary  principles and  concepts  become  further  and  further  removed  from  our  sense  impressions,  and as  a  result  the  claim  that  we  get  them  the  way  empiricists  claim  we  do  becomes  less and  less  credible. A  second  answer  to  the  question  of  how  we  find  these  elements  is  that  they  are  a product  of  pure  reason-that  they  are  a  priori.  This  view  goes  back  at  least  to Ideas  and  Opinions ,  273. 18  Albert  Einstein  and  Leopold  Infeld, The  Evolution  of  Physics:  The  Growth  of  Ideas From  Early  Concepts  to  Relativity  and  Quanta 33;  see  also (New  York:  Simon  and  Schuster,  1942), Ideas  and  Opinions ,  272. Galileo  who  believed  that  keen  insight-what  he  called il  lume  naturale -supplemented  by  a  few  observations  and  experiments,  quickly  yields  them.  Galileo's great  successes  in  dynamics  gave  him  cause  to  be  optimistic,  and  writing  when  he did,  he  could  still  rely  on  a  rather  robust  notion  of  God  as  the  rational  creator  of  the universe  who  in  the  process  created  us  unto  his  own  image.  The  situation  Einstein found  himself  in  is  very  different.  The  notion  of  God  Galileo  relied  upon  was  no longer  plausible  and  the  role  of  choice  in  selecting  primary  principles  and  concepts had  become  much  more  pronounced,  especially  after  the  development  of  non-­Euclidean  geometries  and  alternative  algebras.  Again,  the  result  is  a  view  where  the fundamental  concepts  and  postulates  of  physics  are  in  the  logical  sense  free inventions  of  the  human  mind,  and  they  are  so  without  a  plausible  backstory  on  why or  how  the  mind  can  come  to  comprehend  the  world  that  is  external  to  it. 19  Not  only does  this  mean  that  the  primary  principles  and  concepts  are  fictitious,  but  also  that any  attempt  to  deduce  them  logically  from  our  sense  experiences  is,  in  Einstein's words,  'doomed  to  failure.' 20  That  we  can  comprehend  the  world  appears  on  this view  indeed  utterly  incomprehensible. 3.  Abduction When  calling  a  derivation  logical,  Einstein  is  purely  thinking  in  terms  of  deduction and  induction.  Since  neither  applies  to  the  process  by  which  we  obtain  the  primary principles  and  concepts,  how  we  do  obtain  them  can  thus  only  be  non-­-logical.  'The scientist,'  he  writes  'has  to  worm  [them]  out  of  nature  by  perceiving  in comprehensive  complexes  of  empirical  facts  certain  general  features  which  permit of  precise  formulation.' 21  For  Einstein,  the  primary  principles  and  concepts  are 'freely  chosen  conventions' 22  or  constructions  that  are  the  product  of  a  free  play  in the  imagination,  and  as  there  are  different  ways  of  doing  so,  they  are  'to  a  large 19  The  first  principles  and  concepts  are  a  priori,  but  not  in  the  old  sense  of  being necessary  or  analytic;  they  are  a  priori  only  relative  to  a  theory. Ideas  and  Opinions ,  274. Ideas  and  Opinions ,  221. 22  Albert  Einstein,  'Autobiographical  Notes,'  in  Paul  Arthur  Schilpp  (ed.) Einstein:  Philosopher-Scientist, Albert 2  vols.  (New  York:  Harper  Torchbook,  1959),  1:13. extent  arbitrary.' 23  Our  intuition  plays  a  central  role  here:  'only  intuition,  resting  on systematic  understanding  of  experience,  can  reach  them.' 24  As  such,  Einstein remarks,  they  do  not  sprout  from  a  'deliberate  intention  or  program,'  but  come 'straight  from  the  heart.' 25  Elsewhere  he  describes  'pure  thinking'  in  terms  of playing  the  violin  and  of  smoking  a  pipe  while  lounging  in  an  armchair. 26 This  all  being  said,  Einstein  does  give  some  general  guides  on  how  to  go  about finding  them,  and The  Evolution  of  Physics ,  a  book  he  wrote  with  Leopold  Infeld,  is  in part  an  attempt  to  gain  more  clarity  in  this  matter.  The  suggestion  that  there  might be  rules  raises  the  question  whether  this  decidedly  non-­-logical  process  that  Einstein grounded  in  our  intuition  could  be  recast  in  terms  of  abduction,  the  third  mode  of reasoning  Peirce  distinguished  besides  deduction  and  abduction,  and  if  so  where that  leaves  us.  So  let  us  thus  turn  our  eye  to  Peirce. For  Peirce,  logic  is  a  normative  science:  It  studies  deliberate  thinking  with  the aim  of  distinguishing  good  reasoning  from  bad  reasoning.  Peirce  further  added  that 'to  say  that  any  thinking  is  deliberate  is  to  imply  that  it  is  [self-­-]controlled  with  a view  to  making  it  conform  to  a  purpose  or  ideal.' 27  For  logic,  this  purpose  is  to  attain correct  representations,  which  is  in  line  with  Einstein's  ideas  about  the  purpose  of physics  discussed  earlier.  Peirce  further  maintained  a  naturalistic  account  of reasoning,  arguing  that  it  is  grounded  in  our  problem-­-solving  activity  and  the problem-­-solving  activity  of  many  more  generations  than  that  our  species  is  old. From  this  we  obtained  a  large  toolbox  of  practices  that  serve  us  very  well  in  our day-­-to-­-day  affairs.  It  is  this  toolbox  that  subsequently  becomes  the  subject  of  logic, conceived  as  the  study  of  how  we should reason  as  opposed  to  how  we  actually  do reason.  Now,  since  reasoning  developed  out  of  our  practical  dealings  with  the  world, it  is  far  from  clear  how  well  it  fares  when  applied  to  subjects  that  are  far  removed from  this,  as,  for  instance,  in  theoretical  physics,  where  our  common  sense  or Ideas  and  Opinions ,  300. Ideas  and  Opinions 25  Ibid. 26  Einstein  and  Infeld,  op.  cit.,  4f. 27  Peirce  Edition  Project  (ed.), University  Press,  1998),  376. The  Essential  Peirce, ,  227. Vol.  2  (Bloomington:  Indiana intuition  often  fails  us.  Consequently,  it  is  in  such  areas  that  the  study  of  logic becomes  of  preeminent  importance. In  its  most  general  form,  abduction  can  be  captured  as  follows:  Some  surprising fact B is  observed.  If A were  true,  B  would  be  explicable  as  a  matter  of  course.  So  it  is at  least  plausible  that  A  is  true.  For  instance,  if  we  find  a  bag  of  beans  and  a  cup  of the  same  beans  beside  it,  it  is  reasonable  to  assume  that  the  beans  in  the  cup  came from  the  bag.  Similarly,  the  surprising  fact  of  the  retrograde  motion  of  Mars becomes  a  matter  of  course  once  we  assume  that  Mars  and  Earth  each  orbit  in  an ellipse  that  has  the  sun  at  one  of  its  foci.  In  that  sense  an  abductive  argument  takes the  surprise  out  of  a  surprizing  experience  by  raising  a  plausible  hypothesis  from which  the  experience  follows  as  a  matter  of  course.  As  both  examples  suggest, abduction  is  here  envisioned  in  a  way  that  remains  close  to  the  world  of  sensory observations  and  the  laws  that  can  be  abstracted  from  them,  such  as  Kepler's  first law,  which  is  a  favourite  example  of  Peirce. 28  Consequently,  it  is  not  immediately clear  whether  abduction  would  apply  also  to  the  kind  of  situations  that  Einstein  is talking  of. Even  though  Peirce  explicitly  admitted  that  abduction  'is  very  little  hampered  by logical  rules,' 29 he  still  considered  it  a  type  of  reasoning.  He  did  so  in  part  because  it is  one  of  three  ways  of  combining  rule,  case,  and  result  (deduction  and  induction being  the  other  two)  and  in  part  because  it  is  a  self-­-controlled  deliberate  process  in which  one  tries  to  gain  a  correct  or  at  least  a  plausible  representation,  or,  to  use Einstein's  term,  comprehension.  Peirce  further  used  the  three  ways  of  reasoning  to develop  a  general  recipe  for  science  that  resembles  Einstein's  portrayal  of theoretical  physics:  we  use  abduction  to  establish  an  explanatory  hypothesis,  then use  deduction  to  extract  as  many  consequences  of  it  as  possible,  and  finally  use induction  to  see  whether  these  consequences  match  our  sense  experiences. 28  See  e.g. Peirce Cornelis  de  Waal  (ed.), The  Illustrations  of  the  Logic  of  Science  by  Charles  S. (Chicago,  Open  Court,  2014). 29  Charles  Hartshorne  and  Paul  Weiss  (eds.), The  Collected  Papers  of  Charles  Peirce, Vol.  5  (Cambridge:  Harvard  University  Press,  1934),  sect.  188. Before  exploring  whether  abduction  can  be  used  to  capture  how  Einstein's primary  principles  and  concepts  are  obtained,  we  should  see  why  we  are  so  good  at what  in  essence  comes  down  to  educated  guessing.  Peirce's  answer  is  that  we possess  an  instinct  for  it,  and  he  gives  an  evolutionary  explanation  for  why  this  is  so. He  writes  as  follows:  'if  the  universe  conforms,  with  any  approach  to  accuracy,  to certain  highly  pervasive  laws  or,  in  Einstein's  words,  if  the  universe  is comprehensible],  and  if  man's  mind  has  been  developed  under  the  influence  of those  laws,  it  is  to  be  expected  that  he  should  have  a natural  light, or light  of  nature , or instinctive  insight ,  or  genius,  tending  to  make  him  guess  those  laws  aright,  or nearly  aright.' 30  In  short,  we  possess  reason  because  the  world  does.  From  Peirce's view,  what  makes  the  comprehensibility  of  the  world  incomprehensible  to  Einstein is  that,  like  Galileo,  Einstein  makes  reason  into  something  that  is  independent,  even alien  to  the  world,  and  it  is  this  that  makes  the  successful  application  of  reasoning  to that  world  a  miracle,  especially  when  we  are  reasoning  a  priori. 31  For  Peirce,  in contrast,  abduction  is  possible  because  our  reason  is  the  internalization,  however limited  and  deficient,  of  the  comprehensibility  of  the  universe  itself,  and  accordingly it  becomes  the  task  of  logic  to  make  this  explicit  with  the  aim  of  distinguishing  good from  bad  abductions,  as  we  are  very  good  at  the  latter  too.  Peirce's  account  also suggests  that  abduction  is  more  reliable  when  it  remains  close  to  common  sense  and experience,  as  it  is  within  that  context  that  we  honed  our  talent  for  abduction.  This also  means  that  abduction  might  be  less  trustworthy  when  used  for  the  kind  of  task Einstein  is  talking  of,  especially  given  Peirce's  admission  that  abduction  'is  very little  hampered  by  logical  rules.' So  far  I  described  abduction  as  a  self-­-controlled  deliberate  process.  In  Peirce's view  this  is  not  all  there  is  to  it,  far  from  it.  In  common  perception  we  find  a tremendous  reduction  of  sensory  experience  that  follows  the  logic  of  abduction, 30  Op.  cit.,  sect.  604. 31  It  is  easier  for  Galileo  to  get  away  with  this  then  it  is  for  us  today.  Galileo  could conceive  of  reasoning  as  something  divine  or  supernatural,  while  maintaining  that we-having  been  created  in  His  image-also  partake  in  this  divine  aspect,  even  if  in a  most  deficient  manner;  the  latter  because  our  limitations  as  mortals  and  the various  ways  in  which  our  non-­-rational  elements  impede  and  distort  our  reasoning. even  if  this  reduction  is  largely  if  not  entirely  unconscious. 32  So  perhaps  we  should say  that  abduction,  conceived  as  a deliberate process,  is  a  special  case  of  a  non-­deliberate  process  that  is  far  more  pervasive  and  that  is  the  product  of  the concordant  development  of  mind  and  world.  It  also  means  that  a  situation  where  we arrive  at  a  conclusion  without  knowing  how  we  got  there  can  still  be  an  abductive and  hence  a logical process,  and  this  may  include  the  process  through  which  we  find successful  primary  principles  and  concepts  while  playing  the  violin  or  messing  with our  pipe-that  is,  by  very  deliberately not thinking  about  it.  Moreover,  the justification  of  this  process  lies  entirely  in  how  well  its  products  succeed,  not  in  the method  through  which  they  were  obtained. So  what,  then,  do  we  achieve  by  calling  it  a logical rather  than  a  non-­-logical process?  For  one  thing,  the  Peircean  way  of  looking  at  it  makes  the comprehensibility  of  the  world  again  comprehensible.  Moreover,  subjecting  it  to  the normative  science  of  logic  makes  it  a  process  that  can  be  inquired  into,  at  least  in principle,  and  thereby  it  avoids  the  charge  of  blocking  the  road  of  inquiry,  a  charge to  which  Einstein's  view  is  most  certainly  open.  Thirdly,  it  allows  for  the  formation of  rules,  or  guides,  and  means  for  evaluating  processes. One  might  object  that  by  subjecting  the  process  of  discovery  to  logic  one  is putting  constraints  on  what  should  be  a  free  flow  of  creativity.  There  are  at  least  two responses  to  this.  First,  as  Peirce  remarks,  abduction  is  'very  little  hampered  by logical  rules,'  and  second,  scientists  already  reject  many  fights  of  the  imagination.  If we  cannot  fall  back  on  logic,  as  defined  by  Peirce,  to  determine  which  flights  of  the imagination  are  to  be  taken  seriously  and  which  are  not,  it  seems  we  are  left  with the  intuitions  of  defenders  and  detractors,  prematurely  applied  preconceived notions  (including  those  of  funding  agencies),  existing  power  relations,  and  the  like. Another  objection,  already  hinted  at,  is  that  the  Peircean  notion  of  abduction remains  too  close  to  experience  to  be  truly  applicable  to  the  type  of  situation 32  There  is  even,  Peirce  writes  in  1908,  'some  dose  of  surprise  in  every  perception.' Sheet  9  of  an  unpublished  manuscript  held  in  the  Houghton  Library  at  Harvard University,  numbered  224  by  Richard  Robin  in  his of  Charles  S.  Peirce Annotated  Catalogue  of  the  Papers (Amherst:  University  of  Massachusetts  Press,  1967). Einstein  is  talking  about,  which  is  far  removed  from  the  daily  observations  that  have shaped  our  instinct  for  guessing  right.  In  other  words,  to  treat  the  formation  of hypotheses  as  abductive  is  not  only  too  restrictive,  as  was  first  objected,  but  is  also prone  to  send  us  in  the  wrong  direction  by  suggesting  that  the  entire  universe  must somehow  resemble  the  world  that  we  routinely  experience.  To  this  objection  one may  counter  that  if  our  epistemological  faculty  is  a  function  of  the  comprehensibility of  the  universe,  and  if  it  this  development  took  place  not  just  during  the  age  of science  but  during  the  entire  history  of  our  species  and  all  its  forebears  down  to  its most  primitive  ancestors,  this  faculty  is  more  likely  to  be  reflective  of  the  inner  logic of  the  world  than  to  be  restricted  solely  to  what  can  be  surmised  through  our surface  interactions  with  that  world  through  those  most  remarkable  abductive machines  that  we  have  come  to  call  our  senses. To  conclude,  Peirce's  conception  of  abduction  can  provide  us  with  an  account  of the  comprehensibility  of  the  universe,  and  it  does  so  in  a  naturalistic  way  that avoids  blocking  the  road  of  inquiry.  The  logic  of  abduction  gives  us  a  framework  for deliberately  second-­-guessing  our  guessing  instinct  with  the  aim  of  developing guides  for  the  formation  of  new  hypotheses,  including  those  Einsteinian  primary principles  and  concepts  that  give  us  dynamic  a  priori  frameworks  in  terms  of  which our  sense  experiences  could  be  understood.