the sitters involved    to  assert  that    they    belonged    to  a   particular  class   and its achievements.   Female  sitters
in  these   works   often   addressed   the viewer  through a   direct  outward gaze,   akin    to  a   theatrical  aside,
explicitly  inviting    or  mediating   engagement  in  the construction    of  a   painting’s  meanings    in  such    a   way as
to  draw    attention   to  the act of  looking itself. These   figures were    constructed as  both    viewers (of the
imagined    actual  viewer) and as  objects of  the viewer’s    gaze.   Such    a   device  ran the risk    of  associating the
women   concerned   with    the behaviors   of  acting  or  artifice,   but it  was commonly    used    (see    Figure  2.10).
The hierarchical    classificatory  system  shifted to  accommodate other   genres  that    sat “in between”    official
categories. As  emotional   expressiveness  became  fashionable,    so  did works   known   as  “expressive heads”
(Kayser,    2003,   118).   These   were    general representations of  emotions    (such   as  “Fear”  or  “Innocence”)    in
portrait    formats,    but not based   (or only    very    loosely so) on  specific    individuals.    They    drew    on  stock   types
such    as  the “pupil” or  the “letter writer.”    The late    work    of  Greuze  included    many    such    paintings,  which
appropriated    some    of  the moral   import  and emotional   rhetoric    of  history painting.   This    type    of  painting,
which   combined    a   portrait    format  with    allusions   to  everyday    life    and a   high    expressive  key,    lay
somewhere   at  the intersection    of  portraiture,    history and genre.
The “fantasy    figure,”    which   drew    on  the tradition   of  the “expressive head,”  was another kind    of  work    that
adopted portrait    formats in  order   to  create  imaginative subjects    with    little  or  no  connection  to  identifiable
sitters but presented   as  types   such    as  the “cavalier”  or  the “artist.”   Fragonard   was something   of  a
trendsetter in  this    respect.    His work    evaded  the legible moral   codes   and lessons typical of  history
painting,   especially  when    he  later   worked  independently   of  the Académie    royale  and achieved
commercial  success.    This    is  evident in  the free,   expressive  brushwork   in  which   he  executed    these   works
(Percival,  2012,   38, 50).    The very    term    “fantasy”   was associated  with    rulebreaking    and spontaneous
modes   of  creativity  (Percival,  2012,   12),    although    Fragonard   drew    on  many    precedents  from    earlier art;
for example,    Italian,    Dutch   and Flemish works   of  the seventeenth century and Renaissance such    as  those
by  Rembrandt,  Frans   Hals    (1582/3–1666),Titian    and Caravaggio  (Percival,  2012,   51–63), as  well    as  on
many    more    recent  French  and British works   by  Watteau,    Boucher,    Greuze, Reynolds,   Gainsborough    and
Wright  of  Derby.  The ambiguous   nature  of  Fragonard’s fantasy figures,    and their   emphasis    on  activities
(e.g.   writing,    reading,    painting)   has in  the past    led to  their   classification  as  genre   paintings.  However,    they
seem    to  enter   more    comfortably into    the wide    range   of  subgenres   in  play    in  the eighteenth  century before
the term    “genre  painting”   was in  common  use.    British versions    of  the fantasy figure  appeared    more    rooted
in  everyday    life    (Percival,  2012,   71).    In  general,    the subjects    and meanings    of  fantasy figures were
unclear,    even    when    sitters were    named   in  titles. In  Fragonard’s fantasy figures,    costumes    (often  overtly in
the tradition   of  “dressing   up” or  masquerade),    identities  and expressions were    ambiguous   in  a   way that
was deliberately    intended    to  stimulate   the viewer’s    imagination and to  evoke   alternative,    transgressive
and escapist    identities  and lifestyles, in  a   way that    we  would   now consider    to  be  “modern”    (Barker,
2009,   309–310;    Percival,   2012,   122,    133).
Generic fluidity    worked  very    much    in  favor   of  the “lower” genres, which   gained  in  stature.    In  the case    of
landscape   painting,   there   was a   recognition that    the genre   could   engage  the imagination in  a   way that
reduced its distance    from    history painting.   There   was also    an  increasing  acknowledgment  of  the fact    that    a
wellpainted landscape,  however closely based   on  reality,    was better  than    a   badly   conceived   or
executed    history painting.   Reynolds    revealed,   in  his Discourses, a   brave   departure   in  this    respect:
I   am  well    aware   how much    I   lay myself  open    to  the censure and ridicule    of  the academical  professors
of  other   nations,    in  preferring  the humble  attempts    of  Gainsborough    to  the works   of  those   regular
graduates   in  the great   historical  style.  But we  have    the sanction    of  all mankind in  preferring  genius  in  a
lower   rank    of  art,    to  feebleness  and insipidity  in  the highest.
(Reynolds,  1975    [1797], 249)