|People of The Kalmar Union|
|The Kalmar Union|
Gunnar Atterberg was a mathematician. He is mostly remembered for innovations in the field of automated calculation. Though his own designs would never be completed in his lifetime, his ideas would be copied and refined by others.
He was born in Kvisljungeby, Gothenland in February 1892 to a wealthy banker Jonas Atterberg and, scandalously, his mistress Elizabeth Wiberg. It was only when Gunnar was nine that his father was able to marry Elizabeth and formally recognise him as his son, although he had willingly supported the pair through the years.
Gunnar was quickly recognised as somewhat of a prodigy, especially in mathematics and quickly exhausted the talents of his tutors. Finding the education at the University of Gothenburg only marginally better he drifted through his studies and barely graduated. Teaching jobs both in Gothenberg and around Europe came and went while he built up a collection of friends with interests in various branches of science. Increasingly he was brought in to collate and remove the errors from tables relating to astronomy, insurance and navigation which were reliant on faulty tables. In 1908 he was asked to edit various articles for the Naturventenskapligt Encyclopaedia.
In 1910 whilst working on astromony tables he and his colleague Hendrik Kolind began comparing various published function tables to wheedle out discrepancies. Eventually he came to the conclusion that the tables were full of inaccuracies caused by human calculation, mis-transcribing and poor printing. He knew via the occasional foreign lectures heard in Copenhagen that several nations were trying to solve this problem, mostly by using many mathematicians together (the Aniyunwiyans were apparently using 90 at once) and spliting the calculations into easier parts. Errors would in theory disappear as they were checked over and over. Atterberg proposed something more radical, an automated machine that could calculate the polynomial functions by using finite differences. His ideas would remain on the shelf however.
By 1923 Gunnar was a professor of Mathematics at the University of Gothenburg. Married to an Italian, Juliana Poleni, he already had a large family of six children. He was also independently wealthy thanks to inheriting his father's wealth and position as a 'name' of Jensen's of Copenhagen. Now able to devote more time to personal projects he began setting out plans for his machine, which he dubbed the 'Tabulation Engine'. Realising it would be beyond his own means he petitioned various Kalmar states for funding. Rejected by the larger nations of the West, he would find support from the East.
Duke Leo II of Karelia was eager to exploit the mineral wealth of the duchy as well as raise the education levels of what until recently had been a backward province of Novgorod, and had invested thousands of Markkas in developing a highly skilled education system. A keen amateur mathematician himself the Duke heard of Atterberg's novel solution to the problem of tabulation errors and personally enticed him to leave Gothenburg to join the newly created Academy in Äänislinna as Head of Mathematics. Having been passed over for promotion several times on account of his less than personable demeanor he jumped at the chance. He moved himself and his rapidly growing family to Karelia in the Winter of 1928. Leo II handed him a personal grant of FIM 2,500 and he immediately found an engineering firm, Gaup & Jakobsen, who willing to construct the parts. After a minor delay work began.
The high grade of parts required by Atterberg's machine meant that Gaup & Jakobsen had to redesign many of their tools. Whereas they had previously manufactured parts of steam engines for iron mines they soon became specialists in precision machinery, introducing many improvements to machine lathes and cutters. Construction of the Engine slowly gathered pace but Atterberg continually refined his plans frustrating his partners. Midway through 1934 he produced a radically overhauled design, more powerful but with an automatic printer to remove human error from the printing process. While Gaup & Jakobsen began to put together no. 2 Atterberg toured Scandinavia with no. 1 spurring many inventors and engineers to begin their own experiments in automated calculations.
Atterberg's son Jonas and No.1 would be lost in the shipwreck of the Euippe on 3rd November 1935 and the machine is presumed to be lying off the Bjornholm coast. Atterberg fell into depression mourning the death of his favourite son. Gaup & Jakobsen continued to build no.2 but funding from the Duchy began to dry up as Atterberg withdrew from public life. Leo II's successor Leo III was not interested in the abstract possibility of the machine and looked more to the absolute cost of it. All funding would be cut off in April 1937. No.2, barely a third completed, was finally given to the Academy after costs were settled between Gaup & Jakobsen and Atterberg. The university would continue to slowly add to no.2 as its engineering students were set the yearly challenge of producing parts for it. Gaup & Jakobsen, thanks to the innovations made during no.1 and no.2's construction became leaders in precision engineering and are most famous now for producing parts for artillery pieces for the Kalmar army.
Atterberg's depression led him to abandon the dream of completing his machine. Neglecting his work somewhat he poured his energies into his children nurturing their talents, creativity and ambitions. His sons Per and Olov were supported as they entered Karelian and Gothenlandic politics. Per Atterberg would eventually become mayor of Äänislinna in 1958. Anna-Maria became an accomplished portrait painter. Karine would follow her father's footsteps in automated calculations. Other children such as Lina and Conrad would later complain of their father being sourly over-bearing and a bully. He meanwhile bought a portfolio of mining interests and put them in a trust to provide for his children's future.
Even while construction of no.2 was going ahead Atterberg had been planning something even more ambitious. In early 1935 he had shown plans to both his engineers and the frail Leo II (who had had a stroke in 1933) though did not talk about it on his lecture tours. No.3 was not just simply for calculating polynomial functions, indeed, it was as Atterberg said 'for everything'. Atterberg conceived of it as a series of no.2s linked together with storage and a card reader (such as used in weaving) for 'programming' calculations into the machine. It was reckoned, with 25 digit numbers, to be able to do addition in 3 seconds and multiplication and division in 3 minutes. As the different parts looped around back into the storage unit the machine in theory could be used to complete extremely complex problems. Funding however was simply not available.
The first of Atterberg's children born in Karelia, Karine would help turn her father's designs of no.3 into reality. Karine, by no means Gunnar's favourite child (that position seems to have been taken by Jonas, hence his severe depression on his death) was most like her father in terms of academic achievement. One of the first women to attend the Academy in Äänislinna she graduated with top honours in Mathematics and Chemistry.
Encouraged by the demonstration of Nils Åhlund's simpler (though less accurate) version of a no.2 in 1949 and by reports of other versions being created in France, Anglia and Álengiamark, Gunnar, partially vindicated, appeared to come out of the depression that had stifled his creativity. For the remaining years of his life he would work with Karine to produce theoretical 'programming' for the machine as well as collating the masses of documents he had produced into a coherent plan. She even took over much of the negotiations with the engineers removing the problem of Gunnar's temperamental disposition. In September 1963, one week before his death, the Duchy of Karelia once again began funding efforts to build the machine.
This initial funding had by 1965 produced the 'mylly', the central core of the machine which could do the four regular arithmetic functions. Karine was delighted with its completion but others were less enthusiastic as smaller more practical devices were being produced elsewhere. The Finnish army was interested however and bought the mylly for future development. Without Karine's direct involvement however the project languished. It was passed on to Denmark in 1974 and was officially abandoned three years later†.
Undeterred, Karine would continue to work on various parts, including reproducing the mylly, now on display at the Academy's museum. She would have a hand in developing the aluminium smelting Tiihonen Process as well as establishing a scholarship program for the Academy to support girls without the sufficient funds to afford the fees.
In 1997 she completed a working version of no.2 which, although technically obselete, worked as Gunnar had predicted 70 years before and vindicated his work.
†Although a slip by the Kalmar Secretary of Defense while being questioned over research costs in March 2013 suggests it was not abandoned completely and the special project 'Myren' is nearing completion.