In an earlier blog on Cosmic Diary I published my personal list of the 25 greatest discoveries in astronomy of the 20^th century. See my article of 18 June at http://cosmicdiary.org/blogs/john_hearnshaw/?p=96 on this topic.

In that article I listed my choice of the 25 major discoveries of the last century, and broke the list down into 42 key papers. These papers represented fundamental discoveries of new phenomena, or new types of object, each of which resulted in the development of a major new field of investigation or branch of astronomy. Certainly much subjectivity went into the selection, but I suspect many others who go through this exercise would choose many of the same groundbreaking discoveries, perhaps in a slightly different order.

Attached to these discoveries I have now selected 49 key papers (originally I chose just 42, but in the last few months I have added a few more papers associated with the same 25 discoveries) and these were published by 62 authors. You can see a spreadsheet giving details of these papers and their authors by downloading an Excel file from this website: www2.phys.canterbury.ac.nz/~jhe25/20thC_discoveries1-25.xls.

Key 20th century papers showing the decade of publication (click on figure for enlarged version)

An interesting question is to ask when were these 49 key papers published? The figure shows a histogram according to the decade they appeared. Many were published in the golden decades of 1911-20, in the 1930s and the 1970s, with the ‘50s and ‘60s also being good decades for astronomy. The 1940s were obviously affected by the war years and the first decade was when astrophysics was in its early stages. The surprising point is that relatively few really ground-breaking discoveries were made in the ‘80s and ‘90s.

This is especially remarkable for two reasons. The first is that more than half of all 20th century refereed papers were published since 1981, but only 10% of the key papers were published since that time. If we look at the distribution of publication date of 20^th century refereed papers, there is a dramatic rise from about 13 thousand papers per decade for the first half of the century to as many as 140,000 per decade in the last decade. The histogram shows the results for 446,443 refereed papers in astronomy in the 20^th century, of which 252,519, or 56.6 per cent, were published since 1981.

Total refereed papers published in the 20th century in each decade, showing the more than tenfold rise at the end of the century relative to most earlier decades.

With such a dramatic rise in the publication rate, surely we should expect over half the fundamental breakthrough discoveries to have been since 1981. Moreover, this decline in the rate of key discoveries since 1981 is in spite of the advent of the computer, the CCD, a lot of space astronomy and the opening up of the full electromagnetic spectrum to observations, all accomplished largely since 1981.

I agree that new technology has often led to new discoveries: for example the CCD and fibre optics have made possible the discovery of extrasolar planets using the Doppler method. The development of electronics led to radio-telescopes and to the discovery of the 21-cm line and galactic HI studies. Large optical telescopes have allowed Hubble’s law to be extended to large redshifts and gamma-ray detectors on satellites resulted in the serendipitous discovery of gamma-ray bursters.

But we still have this apparently anomalous result. No matter how you choose the really ground-breaking discoveries, too many of them came before all this amazing new technology, before computers and astronomical satellites and CCDs existed. In recent decades billions of dollars have been poured into astronomy, and this has been why the publication rate has gone up more than tenfold. More money equals more astronomers and more telescopes and other facilities, and these lead to more papers published. But money doesn’t guarantee that the papers will be groundbreaking new discoveries.

My explanation for the startling drop off in the fraction of new papers reporting fundamental ground-breaking discoveries is that astronomers are running out of new types of object or new phenomena to discover. This is not such a bizarre concept. The idea that the number of fundamentally different types of phenomena that exist in the observable universe might be finite seems quite reasonable. Since about 1980 the vast majority of papers are simply adding detail to our knowledge of existing phenomena or types of object.

I certainly agree that astronomy still has much to do; we don’t understand dark matter, or dark energy. We still have to discover truly Earth-like planets in their habitable zones outside the solar system, and we don’t know of any place in the universe other than on Earth where complex life exists. So big discoveries are still awaiting to be made. But to be commensurate with the huge rate of publication and of expenditure in astronomy today, we should expect at least 20 key discoveries per decade if the relative rate of discovery of the golden years of the 20^th century (relative to total papers published) were to be maintained. Could it just be that there are not 20 truly fundamental key phenomena or types of object still remaining to be found in each decade of the current era?

If my explanation is correct, then this could pose a serious threat to the future of research in astronomy. The public has been willing to support mega-science projects like the Hubble Space Telescope because of the huge public outreach program that HST has spawned. But at the end of the day, both the public and astronomers will want to see that taxpayer dollars lead to truly new discoveries, not just reinforcing our knowledge of what is already known.

I am sure this conclusion will be controversial, but I expect that many who have been blinded by the dizzying pace of new papers published and by the beautiful images from Hubble are often those who have not stopped critically to analyse whether today’s papers carry the same momentous importance as the papers of the more distant past. Do today’s discoveries really compare with the discovery of the expansion of the universe (Hubble 1929), or the extragalactic nature of the spiral ‘nebulae’ (Hubble 1925), or the discovery of quasars (Sandage, Schmidt 1963), or the discovery of the nuclear fusion reactions in stars (Bethe 1938), of pulsars (Hewish and Bell 1968), the discovery of the CMBR (Penzias and Wilson 1965) or of gamma-ray bursters (Klebesabel and Strong 1973), or of cosmic radio sources (Jansky, 1933 and Reber 1940) or of cosmic rays (Hess 1912)? There is just a selection of the truly great discoveries of 20^th century astronomy, all of them before computers really became widespread, all before the CCD was used in astronomy and all but one before the advent of astronomy from space. As soon as one begins to select the most important papers from the past, then it becomes apparent that the dearth of key papers of the last 30 years is a really large effect.

Perhaps the situation is akin to the early maritime explorers who were discovering the land masses and geography of the Earth. From the mid-19^th century no major new continents or large islands were discovered. Geographical exploration filled in much detail, and modern aerial mapping means we now know about every corner of the world in exquisite detail. Perhaps the future of astronomy is not so bright after all. Perhaps the public will not fund our science in the future if new discoveries are not made; instead the 21^st century could see a swing to more productive lines of research, such as neuroscience and nanotechnology. I hope I am wrong, but my concerns are at least based on scientific evidence, and those who hold other opinions need to show where my evidence is wrong.

(Written 20 September 2009)

This entry was posted on Thursday, October 8th, 2009 at 12:00 am and is filed under History of astronomy. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.