IELTS Academic Reading Practice Test 263

When the second world war ended in 1945, the economies of Europe were in a worse condition than most people had expected. Factories, ports and railways had been bombed, shipping had been sunk, and millions of people had been displaced from their homes. The continent had also exhausted its financial reserves, and it lacked almost everything needed for reconstruction: coal, machinery, fertiliser and, above all, dollars with which to buy such goods from the United States. In the winter of 1946–47, unusually severe weather and a shortage of fuel closed factories across northern Europe. Even in Britain, which had been on the winning side, food rations were cut.

The American response came in a short speech at Harvard University on 5 June 1947, when the secretary of state, George C. Marshall, received an honorary degree. Marshall said that the United States was willing to help Europe recover, but he insisted that the initiative must come from Europe: the countries of the continent should work out among themselves what they needed and how it should be used. He also denied that the policy was aimed against any country or doctrine, describing its target instead as hunger, poverty and chaos. Nobody in his audience doubted, however, that the spread of communist influence in western Europe lay behind the offer.

Marshall was a soldier by training rather than an economist. He had been chief of staff of the United States army throughout the war, directing the enormous American military effort, and he later served for a year as secretary of defense. His reputation lent weight to a proposal that a less respected figure could never have carried through Congress, and in 1953 he was awarded the Nobel Peace Prize.

European governments responded quickly, but the offer divided the continent. In the summer of 1947 the British foreign secretary, Ernest Bevin, and his French counterpart, Georges Bidault, met in Paris to prepare a European reply and invited the Soviet Union to join them. The Soviet Union sent a large delegation, led by Vyacheslav Molotov, but it soon walked out, and Britain and France then invited other European states to a wider conference to draw up a joint programme. The Americans required the Europeans to plan their needs collectively and to reduce the barriers to trade between them, and the Soviet government would accept neither condition. States in eastern Europe were then instructed to refuse the plan. Czechoslovakia, which had accepted an invitation to Paris, was obliged to withdraw within days.

The scheme still had to be approved by Congress, where many members were reluctant to spend money abroad. A political crisis helped it: in February 1948 the communists took power in Czechoslovakia, and the following month Congress passed the Economic Cooperation Act, which President Truman signed on 3 April 1948. The European Recovery Program, as the scheme was officially called, was run by an agency headed by a businessman, Paul Hoffman. Sixteen countries joined a new body, the Organisation for European Economic Co-operation, which was to share out the aid and press governments to coordinate their investment.

Between 1948 and the end of 1951 the United States supplied goods and money worth roughly thirteen billion dollars. Most of this came as outright grants rather than as loans, and much of it was spent in the United States on American grain, fuel, machinery and fertiliser, which was then shipped across the Atlantic. When European governments sold these goods to their own citizens, they placed the proceeds in special accounts at their central banks. These counterpart funds, as they were known, could be released only with American agreement, and they became an important source of finance for railways, power stations and industry.

Aid came with conditions attached. Governments had to produce four-year plans, report on their progress and begin to dismantle the tariffs, quotas and currency controls that had broken up European trade in the 1930s. American officials argued that a larger and more integrated market would make Europe prosperous, less vulnerable to communism and a better customer for American exporters. Critics at the time, and some historians since, have described the programme as a way of opening Europe to American business as much as an act of generosity.

The recovery that followed was rapid. By 1952 industrial output in western Europe was well above its pre-war level, and trade between European countries had grown sharply. How much of this should be credited to American aid is disputed. Some economic historians have argued that the sums involved were small in relation to the size of European economies, and that recovery was already under way when the first shipments arrived; on this view the plan's main contribution was political and psychological, giving governments the confidence to take difficult decisions. Others reply that without the dollars the shortage of foreign currency would have held back imports of machinery and fuel for years.

American aid ended in 1951, when the programme was replaced by a smaller scheme of military assistance, though some payments continued into the following year. Its institutions outlived it. The Organisation for European Economic Co-operation was later transformed into the Organisation for Economic Co-operation and Development, which still exists, and the habit of cooperation it encouraged is often counted among the foundations of the European Community.

A castle was not simply a fortified building; it was the residence of a lord and the administrative centre of his lands. The form arrived in England with the Norman conquest of 1066, and it spread to Wales, Scotland and Ireland over the following two centuries. The earliest Norman castles were built of earth and timber, often in the motte-and-bailey form, in which a raised mound carried a wooden tower while an enclosed courtyard below held halls, stores and stables. From the late eleventh century these were increasingly rebuilt in stone. Castles were therefore both fortresses and homes, and the balance between the two changed over time. The word itself comes from the Latin castellum, a small fort.

The most imposing part of an early stone castle was the great tower, or keep. Its walls might be three metres thick at the base and were pierced by few openings, while its entrance was usually at first-floor level, reached by an external stair that could be removed in an emergency. The White Tower in London, begun in about 1078, is the best-known example. A keep contained a great hall, a chapel and chambers for the lord's family, along with stores and a well, so that it could be held even if the outer defences fell. Its height was also a statement: a tower visible for miles announced who controlled the surrounding country.

Because a castle had to be entered, the gate was its weakest point, and by the thirteenth century gatehouses had become complex structures in their own right. Attackers approaching the door might first have to pass through a barbican, a walled extension projecting beyond the gate, which forced them into a narrow passage overlooked from both sides. A drawbridge crossed the ditch in front of the entrance and could be raised in an emergency. Behind it an iron or wooden grille, the portcullis, could be dropped in seconds. Above the passage through the gate was a chamber with openings in its floor, through which defenders could drop stones on anyone who had got inside. Guardrooms on either side overlooked the whole approach.

The gate was set in a curtain wall, a continuous ring of stone that enclosed the whole site. Towers were spaced along it and projected outwards, so that archers could shoot at the foot of the wall on either side; a flat wall would have left dead ground where attackers could work in safety. At the top ran a protected walkway, shielded by battlements: a parapet with gaps, called crenellations, through which defenders shot and behind which they reloaded. Narrow vertical slits, the arrow loops, were set in the walls themselves. Beyond the wall lay a broad ditch, often filled with water and known as the moat. Some of the largest castles of the late thirteenth century, including several built for Edward I in north Wales, were given a second ring of walls inside the first, a design known as concentric.

Inside, a castle was a busy household rather than a barracks. The great hall was the centre of public life, where the lord held court, heard complaints and fed his household and guests at long tables. Beyond it were private chambers, kitchens, a brewhouse, a chapel and stables, together with workshops for smiths and carpenters. Furnishings were sparse by modern standards: benches, tables and chests, with hangings on the walls against the cold. Waste from the latrines, or garderobes, usually discharged into a shaft or straight into the ditch. A great castle might house several hundred people, most of them servants, soldiers and craftsmen rather than nobility.

Historians reconstruct such buildings from several kinds of evidence, each with limits. The standing walls themselves show how a castle was built and altered, but many were restored in the nineteenth century, when romantic ideas about the Middle Ages led owners to rebuild and improve ruins. Excavation of ditches, wells and rubbish pits recovers pottery, animal bone and plant remains, which reveal diet and daily life, but only a small proportion of castles has ever been dug. Written records include building accounts, which list wages and materials in detail, and licences to crenellate, royal permissions to fortify a house. Such documents survive mainly for royal works, and few private households kept comparable archives. Scientific dating has also helped: timbers survive in some roofs and floors, and their rings can be matched against known sequences, so that the felling date of the trees can be fixed to the year.

The purpose of all this fortification has been much debated. Sieges were relatively rare, and many castles were never attacked at all, which has led some historians to argue that display and status mattered as much as defence, and that some military-looking features were meant to impress rather than to resist. Others reply that the threat of attack was real even where it never materialised. A familiar claim, that castle stairs were always built to turn clockwise so as to favour right-handed defenders, is now widely doubted, since medieval staircases vary and few were designed with that single aim in mind. Nor did castles simply stop being built: from the fourteenth century, gunpowder artillery made high stone walls increasingly vulnerable, while lords who wanted comfort began to build houses with larger windows and warmer rooms.

Historians have always worked with large quantities of paper, but until recently they have read it one document at a time. The digitisation of archives has changed that. Court records, newspapers, parish registers, tax rolls and ships' logs now exist as searchable text, and software can count, classify and link millions of entries in minutes. Enthusiasts speak of a new kind of history, in which patterns invisible to any individual reader emerge from the data. Sceptics reply that counting is not the same as understanding. Both, in my view, are partly right.

The methods are varied. Text mining extracts names, places and topics from scanned pages; record linkage tries to follow the same person across censuses and registers; statistical analysis of wages, prices and harvests, long established in economic history, now draws on far larger datasets. Literary scholars have described a corresponding shift from close reading of a handful of texts to what has been called distant reading, in which a computer surveys thousands of volumes at once. The results are not summaries but patterns: words that rise and fall in frequency, phrases that cluster in particular decades, groups of documents that resemble one another.

Some of the most striking work has used such sources to recover experience that no chronicle recorded. The digitised Proceedings of the Old Bailey, covering almost two hundred thousand trials between 1674 and 1913, have allowed researchers to trace changes in the language of insult and violence, and a database of transatlantic slave voyages, assembled from shipping records, has documented tens of thousands of crossings. Neither project would have been possible without a computer, and neither was designed to answer the questions it has since been used for.

Promising as these projects are, the archives themselves are not neutral. What survives from the past is what was written down, and most of what was written down was produced by institutions and by people with property, education and power. A machine that reads everything that survives still reads only that. The labourer who never wrote a letter, the woman who never appeared in court and the child who died before being recorded remain as silent in a database of ten million words as in a single manuscript.

Digitisation adds a further filter. Someone has to choose which collections are scanned, and those choices reflect money, fashion and national priorities; newspapers and famous trials are popular, while ledgers and deeds are not. Optical recognition then introduces errors, and the errors are not random: a stained page or a badly printed column may lose exactly the unusual names and archaic words that interest historians. Any claim about long-run change that rests on such material must reckon with the possibility that it is measuring the scanner rather than the past.

A subtler difficulty is classification. To count anything, the historian must decide in advance what categories to use. Occupations must be grouped, spellings standardised, ambiguous entries resolved. Each decision is an interpretation disguised as a technical step, and it may impose modern distinctions on people who would not have recognised them. Linking records is riskier still: two men with the same name in the same town may be one person or two, and a wrong match silently distorts the results. Historians call these false positives, and they multiply as datasets grow.

Beyond these practical problems lies a question of principle. Statistics can show that two things happened together; it cannot say what they meant. The great contribution of historical scholarship has been the patient reconstruction of context, of the assumptions and conventions within which a document made sense. An algorithm that detects a rise in the use of a word has told us nothing until someone asks why speakers and writers reached for it, and that question is answered by reading, not by counting.

Defenders of the new methods have an answer, and it deserves a hearing. Close reading is also selective: a scholar who works through an archive for twenty years still reads a fragment of it, chosen according to instincts that are rarely stated and never tested. Quantitative work, by contrast, makes its assumptions visible, so that others can repeat the analysis and dispute it. Used well, it disciplines intuition rather than replacing it.

My own view is that this argument is too often conducted as though historians had to choose. They do not. The most persuasive work now being done combines the two: a statistical survey that shows where the interesting cases lie, followed by close reading of the documents that explain them, and then a return to the data with better questions. The danger is not the computer but the assumption that whatever can be counted is what matters. Counting only what is easily counted is how the past becomes a dataset and stops being a subject.

History will not become a predictive science, and it should not try. Its business is to explain particular events and the meanings that people gave them, not to forecast the next one. But a discipline that refuses to learn how to handle large collections of evidence will increasingly find that other people, with less understanding of the sources, are handling them instead. The question is not whether big data can produce real history. It is whether historians intend to take part.