INTRODUCTORY

25

shows  that  the  results  reached  on  the  basis
of  a  discriminating  mental  abstraction  (that
is  the  imagined  separation  of  things  that
may  be  inseparable)  hold  with  approximate
accuracy.  A  brain  specialist  may  study  the
brain  alone,  broadly  speaking,  though  no  brain
could  survive  when  severed  from  the  body  with
which  it  was  united.  Similarly  we  may  study
the  economic  activities  of  a  group  of  people,
ignoring  their  religious  and  political  instincts
as  such,  that  is,  so  far  as  they  are  not  expressed
economically,  though  these  instincts  are  an
integral  part  of  human  nature  ;  or,  again,  we
may  study  the  economic  activities  of  one
person,  mentally  separating  him  from  the
community  of  which  he  forms  a  part.
We  may  go  even  further  in  our  defence
of  analytical  economics,  for  there  is  this  very
important  point  to  bear  in  mind,  that  many
of  the  mental  abstractions  that  we  make
as  economists  do  not  assume,  even  for  the
sake  of  argument,  the  isolated  existence
of  things  that  cannot  exist  in  isolation.
Marginal  abstractions  merely  confine  us  to
noticing  how  changes  in  relation  to  a  thing
affect  that  thing.  In  other  words,  thinking
now  of  the  marginal  method  in  economics
merely,  our  abstractions  consist  merely  in
focusing  attention  on  a  thing  and  difieren-