THENEWYORKER,NOVEMBER29, 2021 65
mega-junctions as well as some awk-
ward dead ends. Jim Rossignol and I
reached the goal in nine underwhelm-
ing minutes, after barely even trying.
“It feels like a seventeen-minute guitar
solo railroaded into a four-minute pop
song,” Rossignol said. So, rather than
take advantage of Fisher’s escape valve,
we decided to deliberately enmaze our-
selves, venturing into more feral regions,
where we were soon completely alone.
Greener, taller hedges encroached on
already cramped, mossed-over paths,
and clusters of ripe blackberries poked
through the yew walls. Around one cor-
ner, a wood pigeon landed, gave us a
hard stare, and flew away. “Perhaps the
maze is a sorting mechanism?” Rossig-
nol suggested. “Only those who want
to get lost do.”
As we wandered around in circles for
nearly an hour, Rossignol explained that
the question of difficulty is controver-
sial in the gaming community. “Should
you have to get good at a game to enjoy
it?” he said, summarizing the terms of
the debate. “Or should there be an easy
mode?” These days, game designers study
heat maps to see how players move
through a virtual maze. Just as mainte-
nance of the less trafficked sections of
Longleat’s maze had clearly lapsed, pro-
grammers will often streamline their
virtual environments accordingly, de-
cluttering a game’s memory budget.
Saward sees Longleat as represent-
ing a moment of transition between
modern, Fisher-style puzzles that are
optimized for entertainment and the
less user-friendly, more atmospheric
mazes of the past. “A lot of modern
mazes are designed with an eye to how
many people can we get through the
gate per hour and how quickly can we
get them to the gift shop,” he explained.
“Longleat is a bit old school.”
I
cannot recommend attempting an
Adrian Fisher maze in the company
of its designer. At Escot, a country house
and estate in Devon, we spent a full hour
lost within a hedge maze just half the
size of an American-football field. On a
Wednesday afternoon in early Septem-
ber, a handful of families were also
crunching through its wide gravel paths,
between well-maintained beechwood
hedges that turn from verdant lime green
to autumnal rust as the seasons change.
Small kids raced ahead in twos and threes,
shouting confidently that they knew the
way. Their mothers and grandmothers
moved at a more leisurely pace, carrying
sweaters and snacks. And, as I became
hopelessly lost, Fisher provided a teas-
ing commentary in the old-fashioned
tones of a zany children’s-TV presenter.
“Which way should we go?” he asked
at a four-way junction under a pergola.
“I don’t know! We are spoiled for choice.”
As I returned to the entrance for the
third time in thirty minutes, he asked
me whether I’d noticed that we’d prob-
ably already walked twice the length of
the maze. And, when I foolishly dared
to speculate that I’d finally found the
right path, only to end up on the other
side of the hedge from the goal, he gaily
pretended to commiserate: “Who would
have guessed? What a beastly joke!” Most
enraging of all, he swore that he also had
no idea of the correct route, having de-
signed the maze way back in 2004. Still,
I couldn’t help but feel that I was being
tested, and, to Fisher’s evident delight,
failing. (When I told his brother Bill
about this adventure, he laughed and
said, “It depends on how he was feeling
on the day, whether he actually got lost
or whether he was having fun with you—
and you’ll never know, because you’ll
never get a straight answer.”)
Fisher’s design at Escot combines
several features intended to create max-
imum disorientation for the minimum
cost. At their most basic, all hedge mazes
rely on turns and high hedges to disrupt
the kind of landmark-based navigation
on which humans rely. The homoge-
nous nature of maze materials—gravel
paths and beech leaves, all shadowless
under the diffuse light of a cloudy En-
glish sky—is important, too. Paul Dud-
chenko, a behavioral neuroscientist at
the University of Stirling and one of a
handful of researchers in the world who
study getting lost, told me that our sense
of direction often fails in environments
that are “difficult to disambiguate”—for-
ests, say, or airports.
Fisher had added another layer of am-
biguity, by placing a number of identical
square wrought-iron pergolas at four-
way junctions. “If the design has more
than one of these, you may mistakenly
think you’re in the same place, when in
fact you’re in a different place,” Fisher
warned, as we emerged from a hedge cor-
ridor and encountered one. Because the
paths to the pergolas meet at right angles,
and because Escot’s maze is square, peo-
ple arriving at a pergola tend to assume
that its edges are parallel to those of the
maze. They are unlikely to suspect that
the pergola is one of five, positioned in
pentagonal symmetry around the center
of the maze—an orientation inherently
confounding to our anatomically derived
propensity to think in terms of forward,
backward, left, and right. Even aspirants
who happen to know that there are five
pergolas have to remember not just which
of the three other paths they have cho-
sen before but which one of the twenty
such choices they face in total.
“Options are opponents,” Bret Roth-
stein, an art historian who studies puz-
zles, told me. Fisher’s design, he ex-
plained, deliberately “weighs one down
with options, each of which creates fur-
ther options, while closing off others.”
Many of those pergola paths led around
in a loop, straight back to the same per-
gola. “You made your choice, and, blow
me down, it was just two sides of a sin-
gle hedge,” Fisher said, as he followed
behind me, narrating my wanderings.
Elsewhere in the maze, there were
long stretches without any junctions.
Oskar van Deventer, a Dutch telecom
engineer and a renowned designer of
mechanical puzzles, told me, “This is
something you will recognize in all
Adrian Fisher mazes: that it has some
long corridors with no decision to be
made.” This provides the choice-fatigued
aspirant with a brief, blissful break, but,
of course, as I discovered when I hit one
and thought I must finally be on the
right track, it also serves Fisher’s wily
purposes. “A long journey with no choices
reinforces the feeling that either you’re
going to solve it—or you’re getting very
lost,” Fisher explained.
Escot’s bridges are similarly mislead-
ing: I approached my first with a sense
of relief, only to discover that they offer
just enough vertical perspective to make
you think you can plan your route but
too little to actually figure out the whole
maze. “It’s sort of, like, Let me give you
Hedge mazes rely on turns and high hedges to disrupt the kind of landmark-based navigation on which humans rely.
