FINAL TEST FOR UPPER-INTERMEDIATE UPPER-INTERMEDIATE CLASSES READING MODULE Reading Passage 1 You should spend about 20 minutes on Questions 1-13, which are based on Reading Passage 1 below.
Walking wit din!sa"#s Peter L. Falkingham and his colleagues at Manchester University are are deve develo lopi ping ng tech techni niu ues es !hic !hich h
look look set set to revo revolu luti tion onis ise e our our
understanding o" ho! dinosaurs and other e#tinct animals $ehaved. The media image of palaeontologists who study prehistoric life is often of eld workers camped in the desert in the hot sun, carefully picking away at the rock surrounding surrounding a large la rge dinosaur bone. But Peter Falkingham Falkingham has done little of that for a while now. Instead, he devotes himself to his computer. ot because he has become inundated with paperwork, but because he is a new kind of palaeontologist! a computational palaeontologist. "hat "hat few few peop people le may cons consid ider er is that that unco uncove veri ring ng a skel skelet eton on,, or discovering a new species, is where the research begins, not where it ends. "hat "hat we really eally want want to unders understan tand d is how the e#tinct e#tinct animal animals s and plants plants behaved in their natural habitats. $rs Bill %ellers and Phil &anning from the 'niversity of &anchester use a (genetic algorithm() a kind of computer code that that can can chan change ge itse itself lf and( and(ev evol olve ve() ()to to e#plo #plorre how how e#tin #tinct ct anim animal als s lik like dinosaurs, and our own early ancestors, walked and stalked. stalked. The fossilised bones of a complete dinosaur skeleton can tell scientists a lot about the animal, but they do not make up the complete picture and the comp comput uter er can can try try to ll ll the the gap. gap. The The comp comput uter er mode modell is give given n a digi digiti tise sed d skele skeleton ton,, and the locati locations ons of known known muscl muscles. es. The The model model then then rando randomly mly activates the muscles. This, perhaps unsurprisingly, results almost without fail in the animal falling on its face. %o the computer alters the activation pattern and tries again ... usually to similar e*ect. The modelled (dinosaurs( +uickly (evolve(. If there is any improvement, the computer discards the old pattern
%T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
1
and adopts the new one as the base for alteration. ventually, the muscle activation pattern evolves a stable way of moving, the best possible solution is reached, and the dinosaur can walk, run, chase or gra8e. 2ssuming natural selection evolves the best possible solution too, the modelled animal should be moving in a manner similar to its now)e#tinct counterpart. 2nd indeed, using the same method for living animals 9humans, emu and ostriches: similar top spee sp eeds ds wer were achi achiev eved ed on the the comp comput uter er as in reali eality ty.. By comp compar arin ing g thei theirr cyberspace results with real measurements of living species, the &anchester team of palaeontologists can be condent in the results computed showing how e#tinct prehistoric animals such as dinosaurs moved. The &anchester 'niversity team have used the computer simulations to produce
a
model
of
a
giant
meat)eating
dinosaur.
It
is
called
an
acroca acrocanth nthosa osauru urus s which which litera literally lly means means (high (high spi spined ned li8ar li8ard( d( becaus because e of the spines which run along its backbone. It is not really known why they are there but scientists have speculated they could have supported a hump that stored fat and water reserves. There are also those who believe that the spines acted as a support for a sail. 0f these, one half think it was used as a display and could ould be ;us ush hed wit with blo blood and and the the othe otherr half alf thin think k it was used as a temperature)regulating device. It may have been a mi#ture of the two. The skull seems out of proportion with its thick, heavy body because it is so narrow and the
Falkingham uses computational techni+ues to model a volume of mud and control the moisture content, consistency, and other conditions to simulate the mud of prehistoric times. 2 footprint is then made in the digital mud by a virtual foot. This footprint can be chopped up and viewed from any angle and str stress ess valu values es can can be e#tr e#trac acte ted d and and calc calcul ulat ated ed from from insi inside de it. it. By runn runnin ing g hundreds of these simulations simultaneously on supercomputers, Falkingham can start to understand what types of footprint would be e#pected if an animal moved in a certain way over a given kind of ground. ooking at the variation in the virtual tracks, researchers can make sense of fossil tracks with greater condence. The
application
of
computational
techni+ues
in
palaeontology
is
beco becomi ming ng mor more prev preval alen entt ever every y year year.. 2s comp comput uter er powe powerr cont contin inue ues s to increase, the range of problems that can be tackled and +uestions that can be answered answered will only e#pand. Questions 1-% $o the follow following ing statem statement ents s agree agree with with the infor informat mation ion given given in /eading eading Passage > ? 'n boxes 1-6 on your answer sheet write &'U(
i! the statement agrees with the in!ormation
F)L*(
i! the statement contradicts the in!ormation
+& /(+ >
i! there is no in!ormation on this
In his study study of prehist prehistoric oric life, Peter Peter Falkingham alkingham rarely rarely spends spends time on outdoor research these days.
@
%evera %everall atte attemp mpts ts are are usu usuall ally y need needed ed befor before e the the comp compute uterr mode modell of of a dinosaur used by %ellers and &anning manages to stay upright.
4
"hen the %ellers %ellers and &anning &anning computer computer model model was used for people, people, it showed them moving faster than they are physically able to.
6
%om %ome
palae alaeo ontol ntolog ogis ists ts
have ave
e#pr #presse essed d
reser eserv vatio ations ns
abou aboutt
the the
conclusions reached by the &anchester team concerning the movement of dinosaurs. A
2n e#perien e#perienced ced tracker tracker can analyse analyse fossil fossil footprint footprints s as easily easily as those those made by live animals.
/esearch esearch carried carried out into the composit composition ion of prehist prehistoric oric mud has been found to be inaccurate.
Questions 0- %T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
3
"abel the diagram below. #hoose $% &%R ()*$ %$ +%R, !rom the passage !or each answer. +rite your answers in boxes - on your answer sheet.
4
Ques Questi tion ons s 12-13 12-13 #omplete #omplete the /ow-chart /ow-chart below. below. +rite ite + M'( &)+ &4 4'5*.
%T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
5
Reading Passage $ You should spend about 20 minutes on Questions 16-7%, which are based on Reading Passage 2 below.
Te #!%!ts a#e &!'ing - !# a#e te() +hat is the current state o! play in *rticial ntelligence 2
Can robots robots advance advance so far that they become become the ultimate ultimate threat threat to our
e#ist e#istenc ence? e? %ome %ome scient scientist ists s say no, no, and dis dismis miss s the very very idea idea of 2rtic 2rticial ial Intelligence. The human brain, they argue, is the most complicated system ever created, and any machine designed to reproduce reproduce human thought is bound to fail. Physicist /oger Penrose of 0#ford 'niversity and others believe that machines are physically incapable of human thought. Colin &c1inn of /utgers 'niversity backs this up when he says that 2rticial Intelligence (is like sheep trying to do complicated psychoanalysis. They
B
2rticial 2rticial Intelligen Intelligence, ce, or 2l, is di*erent di*erent from from most most technolog technologies ies in that
scientists still understand very little about how intelligence works. Physicists have a good understanding of ewtonian mechanics and the +uantum theory of atoms atoms and and molec molecule ules, s, where whereas as the basic basic laws laws of intell intellige igence nce rema remain in a
6
mystery. But a si8eable number of mathematicians and computer scientists, who are specialists in the area, are optimistic about the possibilities. To To them it is only a matter of time before a thinking machine walks out of the laboratory. 0ver the years, various problems have impeded all e*orts to create robots. To attack attack these these di= di=cul cultie ties, s, resea researc rcher hers s tried tried to use the (top)d (top)dow own n appro approach ach(, (, using a computer in an attempt to program all the essential rules onto a single disc. By inserting this into a machine, it would then become self)aware and attain human)like intelligence. C
In the the >DA >DAEs Es and and >D >DEs Es gr great eat prog progrress ess was was made made,, bu butt the the shor shortc tcom omin ings gs
of these prototype robots soon became clear. They were huge and took hours to navigate across a room. &eanwhile, a fruit ;y, with a brain containing only a fraction of the computing power, can e*ortlessly navigate in three dimensions. 0ur 0ur brai brains ns,, lik like the the frui fruitt ;y(s ;y(s,, unco uncons nsci ciou ousl sly y recog ecogni nise se what what we see see by performing countless calculations. This unconscious awareness of patterns is e#actly what computers are missing. The second problem is robots( lack of common sense. umans know that water is wet and that mothers are older than than their their daught daughters ers.. But there there is no mathe mathema matic tics s that that can e#pr e#press ess these these truths. Children learn the intuitive laws of biology and physics by interacting with the real world. /obots know only what has been programmed into them. $
Beca Becaus use e of the limit limitat atio ions ns of the the top) top)do down wn appr approach oach to 2rti 2rtic cia iall
Intelligence, attempts have been made to use a (bottom)up( approach instead ) that is, to try to imitate evolution and the way a baby learns. /odney Brooks was was the the dir directo ectorr of &IT( &IT(s s 2rti 2rtic cia iall Inte Intell llig igen ence ce labo labora rato tory ry,, famo famous us for for its its lumbering (top)down( walking robots. e changed the course of research when he e#plored the unorthodo# idea of tiny (insectoid( robots that learned to walk by bumpin bumping g into into things things instea instead d of comput computing ing mathe mathema matic ticall ally y the preci precise se posit position ion of their their feet. feet. Today oday many many of the descen descendan dants ts of Brook Brooks( s( insect insectoid oid robots are on &ars gathering data for 2%2 9The ational 2eronautics and %pace 2dministration:, 2dministration:, running across the dusty landscape of the planet. For all their successes in mimicking the behaviour of insects, however, robots using neural networks have performed miserably when their programmers programmers have tried to duplicate in them the behaviour of higher organisms such as mammals. &IT(s &arvin &insky summarises the problems of 2l! (The history of 2l is sort of funny funny becaus because e the rst rst real accom accompli plishm shment ents s were were beauti beautiful ful things things,, like like a machine that could do well in a maths course. But then we started to try to %T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
7
make machines that could answer +uestions about simple children(s stories. There(s no machine today that can do that.(
There are people who believe that eventually there will be a combination
between the top)down and bottom)up, which may provide the key to 2rticial Intell Intellige igence nce.. 2s adults adults,, we blend blend the distingu distinguish ishes es us as human, human, that that it is impo imposs ssib ible le for for mach machin ines es ever ever to have have emot emotio ions ns.. Comp Comput uter er e#pe e#pert rt ans ans &oravec thinks that in the future robots will be programmed with emotions such as fear to protect themselves so that they can signal to humans when their their batter batteries ies are running low, for e#ample. motions are vital in decision) making. People who have su*ered a certain kind of brain in
Ther There e is no unive univers rsal al cons consen ensu sus s as to whet whethe herr mach machin ines es can can be
cons consci ciou ous, s, or even even,, in hum human ter terms, ms, what what cons consci ciou ousn snes ess s mean means. s. &ins &insk ky suggests the thinking process in our brain is not localised but spread out, with di*erent centres centres competing with one another at any given time. Consciousness may then be viewed as a se+uence of thoughts and images issuing from these di*erent, smaller (minds(, each one competing for our attention. /obots might eventu eventuall ally y attain attain a (si (silic licon on consci conscious ousnes ness(. s(. /obots, obots, in fact, fact, might might one day embody an architecture for thinking and processing information that is di*erent from ours ) but also indistinguishable. If that happens, the +uestion of whether they really (understand( becomes largely irrelevant. 2 robot that has perfect mastery of synta#, for all practical purposes, understands what is being said. Questions 16-72 /eading Passage @ has si# paragraphs, *-3 paragraphs, *-3.. "hich paragraph contains the following information? +rite the correct letter letter *-3 *-3 in boxes 14-20 on your answer sheet. $5 You may use any letter more than once. >6
an insect that proves the superiority of natural intelligence over 2rticial
Intelligence >A
robots being able to benet from their mistakes
8
>
many many resear researchers chers not being put o* believing believing that 2rticial 2rticial Intelligen Intelligence ce will eventually be developed
>G >5
an innovative approach that is having limited success the possibilit possibility y of creating creating 2rticial 2rticial Intelligen Intelligence ce being doubted doubted by some
academics >D @E
no generally accepted agreement agreement of what our brains do robots robots not being able to e#tend e#tend their intelligen intelligence ce in the same way as
humans
%T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
9
Questions 71-73 "oo "oo at the !ollow !ollowing ing people people 78uest 78uestion ions s 21-29: 21-29: and the lis listt o! state statemen ments ts below. &atch each person with the correct statement *-. +rite the correct letter letter *- in boxes 21-29 on your answer sheet. @>
Colin &c1inn
@@
&arvin &insky
@4
ans &oravec 2
2rticial Intelligence may may re+uire re+uire something something e+uivalent to feelings feelings in order to succeed.
B
$i*erent $i*erent kinds of people use di*erent parts of the brain.
C
Tests involving involvin g ction have defeated defeate d 2rticial Intelligence Intelli gence so far. far.
$
People have intellectual capacities which do not e#ist in computers.
People have no reason to be frightened of robots.
Questions 76-7% #omplete the summary below. #hoose +( 4'5 +L8 !rom the passage !or each answer. +rite your answers in boxes 24-26 on your answer sheet. "hen will we have a thinking machine? $espite $espite some some advances, advances, the early early robots robots had certain certain weakness weaknesses. es. They were given the information they needed on a @6.............This was known as the (top) down( approach and and enab enable led d them them to do cert certai ain n task tasks s but but they they wer were unab unable le to recog ecogni nise se @A........ @A............. .....or or did they have any intuition intuition or ability to make decisions decisions based based on e#perien e#perience. ce. /odney Brooks Brooks tried a di*eren di*erentt approach. approach. /obots /obots similar similar to thos those e inve invent nted ed by Broo Brooks ks are are to be foun found d on @... @..... .... .... .... ...w .whe herre they they are are collecting information.
Reading Passage * You should spend about 20 minutes on Questions 70-62, which are based on Reading Passage 9 below.
10
Endange#ed lang"ages '$e; '$e;er er mind ind whal whales es sa;e sa;e the the lang langua uage ges' s' says says Pete Peterr &ona &onagh ghan an graduate o! the *ustralian $ational DGEs to support an initiative that has resulted in good documentation of most of the >4E remai remainin ning g 2borig 2borigina inall langua languages ges.. In nglan ngland, d, anothe anotherr 2u 2ustr strali alian, an, Peter eter 2ustin, has directed one of the world(s most active e*orts to limit language loss, at the 'niversity of ondon. 2ustin heads a programme that has trained many documentary linguists in ngland as well as in language)loss hotspots such as "est 2frica and %outh 2merica. 2t linguistics meetings in the '%, where the endangered)language issue has of late been something of a ;avour of the month, there is growing evidence that not all approaches to the preservation of languages will be particularly help helpfu ful. l. %om %ome ling lingui uist sts s are are boas boasti ting ng,, for for e#amp #ample le,, of more ore and and mor more
%T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
11
sophisticated means of capturing languages! digital recording recording and storage, and internet and mobile phone technologies. But these are encouraging the (+uick dash dash(( styl style e of recor ecordi ding ng trip trip!! ;y in, in, swit switch ch on digi digita tall recor ecorde der, r, ;y home home,, download to hard drive, and store gathered material for future research. That(s not +uite what some endangered)language specialists have been seeking for more than 4E years. &ost loud and untiring has been &ichael Krauss, of the 'niversity of 2laska. e has often complained that linguists are playing with non)essentials while most of their raw data is disappearing. "ho is to blame? That prominent linguist oam Chomsky, say Krauss and many others. 0r, more precisely, they blame those linguists who have been obsessed with his approaches. inguists who go out into communities to study, document and describe languages, argue that theoretical linguists, who draw conclu conclusio sions ns about about how how langua languages ges work, work, have have had so much much in;uen in;uence ce that that linguistics has largely ignored the continuing disappearance of languages. Chomsky, from his post at the &assachusetts Institute of Technology, has been the great man of theoretical linguistics for far longer than he has been known as a political commentator. is landmark work of >DAG argues that all langua languages ges e#hib e#hibit it certai certain n univer universal sal gramm grammati atical cal featur features, es, encode encoded d in the huma human n mind mind.. 2mer 2meric ican an ling lingui uist sts, s, in part partic icul ular ar,, have have focu focuse sed d lar largely gely on theo theorretic etical al conc concer erns ns ever ever sinc since, e, even even whil while e doub doubts ts have have moun mounte ted d abou aboutt Chomsky(s universals. 2ustin and Co. are in no doubt that because languages are uni+ue, even if they do tend to have common underlying features, creating dictionaries and gram gramm mars ars
re+ui e+uirres
prol prolon onge ged d
and and
dedi dedica cate ted d
work work..
This This re+ui e+uirres
that that
documentary linguists observe not only languages( structural subtleties, but also related social, historical and political factors. %uch work calls for persistent funding of eld scientists who may sometimes have to venture into harsh and even even ha8a ha8arrdous dous plac places es.. 0nce 0nce ther there, e, they they may face face di= di=cult cultie ies s su such ch as com communi munity ty su susp spic icio ion. n. 2s ick ick van vans s says says,, a comm commun unit ity y who who sp spea eak k an endangered language may have reasons to doubt or even oppose e*orts to preserve it. They may have seen support and funding for such work come and go. They may have given up using the language with their children, believing they will benet from speaking a more widely understood one. Plen Plenty ty of stud studen ents ts cont contin inue ue to be draw drawn n to the the inte intell llec ectu tual al thri thrill ll of linguistics eld work. That(s all the more reason reason to clear away barriers, barriers, contend 12
vans, 2ustin and others. The highest barrier, they agree, is that the linguistics profe professi ssion( on(s s emphas emphasis is on theor theory y gradu graduall ally y wears wears down down the enthus enthusias iasm m of linguists who work in communities. Chomsky disagrees. e has recently begun to speak in support of language preservation. But his linguistic, as opposed to humanitarian, argument is, let(s say, unsentimental! the loss of a language, he states, (is much more of a tragedy for linguists whose interests are mostly theo theorretic etical al,, lik like me, me, than than for for ling lingui uist sts s who who focu focus s on desc descri ribi bing ng sp spec eci ic c languages, since it means the permanent loss of the most relevant data for general theoretical work(. 2t the moment, few institutions award doctorates for such work, and that(s the way it should be, he reasons. In linguistics, as in ever every y othe otherr disc discip ipli line ne,, he beli believ eves es that that good good desc descri ript ptiv ive e work work re+ui e+uirres thorough theoretical understanding and should also contribute to building new theor theory y. Bu Butt that(s that(s preci precisel sely y what what docum document entati ation on does, does, ob
i! the statement agrees with the ;iews o! the writer
+
i! the statement contradicts the ;iews o! the writer
+& /(+
i! it is impossible to say what the writer thins about this
@G @5
By @EAE only a small number of languages will be ;ourishing. 2ustralian academics( e*orts to record record e#isting 2boriginal languages have been too limited.
@D
The use of technology in language research research is proving unsatisfactory in
some respects. 4E 4> 4@
Chomsky(s Chomsky(s political views have overshadowed his academic work. $ocumentary linguistics studies re+uire re+uire long)term nancial support. Chomsky(s Chomsky(s attitude to disappearing languages is too emotional.
Questions 33-3%9 #hoose the correct letter * 5 # or ,. 44
The writer mentions rainforests rainforests and the o8one layer
%T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
13
2
becaus because e he believ believes es an#iety an#iety about about environm environment ental al issues issues is unfoun unfounded ded..
B
to demonstrate that academics in di*erent disciplines share the same
problems. C
because because they e#empl e#emplify ify what is wrong wrong with the attitudes attitudes of some some
academics. $
to make the point that the public should be e+ually concerned about
languages. 46
"hat does ick vans say about speakers speakers of a Creole?
2
They lose the ability to e#press e#press ideas which are part of their culture.
B
0lder 0lder and younge youngerr member members s of the commu communit nity y have have di= di=cul culty ty
communicating. C
They e#press e#press their ideas more clearly and concisely than most people.
$
2ccessing practical information information causes problems problems for them.
4A
"hat is similar about "est 2frica and %outh 2merica, from the linguist(s
point of view? 2
The nglish language is widely used by academics and teachers.
B
The docum document entary ary lingui linguists sts who work work there there were were traine trained d by
2ustralians. C
ocal languages are disappearing rapidly in both places.
$
There are now only a few undocumented languages there.
4
&ichael Krauss has fre+uently pointed out that
2
linguists are failing to record record languages before they die out.
B
linguists linguist s have made poor use of improvements in technology technolo gy..
C
linguistics has declined in popularity as an academic sub
$
linguistics departments are underfunded in most universities.
Questions 30-62 :omplete each sentence !ith the correct ending, )-, $elo!. 4rite the correct letter, )-, in $o#es 30-62 on your ans!er sheet. 4G
inguists like Peter 2ustin believe that every language is uni+ue
45
ick ick vans vans su sugge ggests sts a commu communit nity y may resis resistt attemp attempts ts to save save its
language 4D
&any young researchers researchers are interested in doing practical research research
6E
Chomsky Chomsky supports work in descriptive linguistics 2
even though it is in danger of disappearing.
B
provided that it has a strong basis in theory. theory. 14
C
although it may share certain universal characteristics.
$
because there is a practical advantage to it.
so long as the drawbacks are clearly understood.
F
in spite of the prevalence of theoretical linguistics.
1
until they realise what is involved.
%T - IT% P/01/2& F0/ 03/%2% %T'$I% - Phone! 4565 6644 - "eb! set)edu.com7web
15
