THE MECHANICAL
FALLACY 107
his canvas orcompounds his
pigments, butwe do
take
delightintheadjustmentofsupporttoload,and
thrust to thrust. Itisno doubtlegitimate toadd
decorativedetail tothesefunctionalelements
;
they
may be enriched
by colour or
carving
;
but our
pleasureinthecolourandthecarvingwillbepleasure
in painting
or sculpture
;
our specifically
architec-
turalpleasurewillbeinthefunctionsofthestructural
elements
themselves. Itisinthisvividconstructive
significance ofcolumnsandarches thattheirarchi-
tecturalbeauty lies, andnotsimplyin theircolour
andshape,assuch,andsofarasthestructuralvalues
areabsent,andthe eyeis merelycharmedbyother
qualities, it is no
longer architectural beautythat
weenjoy. Only,thesefunctional elementsmust
be
vividly
expressed,
and, ifnecessary, expressed with
emphasisandexaggeration. Thesupporting
members
must assure us of theirsupport. Thus, the Doric
or the Romanesquemassiveness, while
it wasin a
sense bad science, was good art
;
yet its beauty
wasnone thelessessentiallystructural.
Thus, the
printingpressortherailwaystationwillnowappro-
priatelyfall outside our definitionbecause,
although
truthfullyand
perfectlyconstructed,andfitfor
their
functions, theydo not
vividlyenoughexpress.what
thosefunctions are,northeir fitnessfor
performing
them. Structurally
perfect, theyare still structu-
rallyunbeautiful. Ontheotherhand,
thearchesand