by Yong Zhao
What do you do when you don't care about something but have to deal with it? Spending a few hundred dollars to buy a publication strikes many Chinese professionals as a reasonable, albeit unethical, choice. It is unlikely that a paper purchased online will qualify for important science journals, especially those published in English outside China. And papers published in fake journals are not counted as China's total paper output. However, the mechanism that drives Chinese professionals to purchase papers and false publications also fuels the production of legitimate publications. If the mechanism is powerful enough to motivate so many professionals to commit fraud, it is undoubtedly powerful enough to motivate millions of professionals to do real research and publish it. With such a large population of individuals driven to produce research papers, it should not be surprising that some of them make it into SCI-indexed journals, which vary in quality and selectivity. The 116,000 papers published by Chinese researchers in 2012 represent only a sliver (about 1 percent) of the total number of papers produced by the tens of millions of professionals required to publish each year.
It Pays to Publish
Sticks work; so do carrots. While many Chinese professionals are driven to produce papers in order to keep their job or seek promotion, others are driven by the prospect of extra cash or perks.
In 2009, Jinggangshan or Jinggang Mountains, the first revolutionary base established by Mao Zedong and his Communist comrades nearly a century ago, became the center of national and even international attention again, this time for trying to fulfill the wish of the Chinese government to advance science and technology. A university is now located in—and named after—the mountains. In just two years, two of Jinggangshan University's professors published seventy papers in the UK-based journal Acta Crystallographica Section E. Alas, these papers were retracted in 2009 because they had reported fabricated and falsified data. The journal later announced more retractions. Although the thirty or so papers retracted later were not authored by these same two professors, most of them came from researchers at the same university.
Further investigation undertaken by Chinese journalists revealed that virtually all of the seventy retracted papers were authored by one researcher, Zhong Hua, who had been recruited to the university with a master's degree in 2004. Zhong discovered a quick way to rack up international publications through Acta Crystallographica Section E, a legitimate open-access online journal that was once included in SCI.14 However the journal is considered a megajournal, a crossover between a traditional scientific journal and a database. Its role is to publish reports of new crystal structures, so if a structure is new, it gets published. The journal has been a major outlet for Chinese researchers in this field. A Nature report published in 2010 says that “half of the 200,000-odd crystal structures published by the journal during the past five years have come from China.”15
With forty-one papers published in the journal between 2006 and 2008, Zhong Hua had been quite successful. He was generous with that success, sharing it with his friends. One day his wife, who came from the same city as the wife of another researcher, Liu Tao, told her that Zhong Hua could help her husband get SCI publications. Liu accepted the offer, even though his research field had nothing to do with identifying crystal structures. In the end, he had twenty-nine papers published under his name but produced by Zhong.
For his forty-one papers, Zhong would have received roughly 200,000 yuan (about $30,000) in cash prizes, twice as much as the average college professor's annual salary. A 5,000 yuan cash prize was awarded to each SCI paper publication.16
To motivate research productivity, Chinese universities, research organizations, and governments have devised an elaborate system that gives out generous cash prizes to researchers for their publications, patents, contracts, and grants. The amount may differ across institutions, but the system follows a similar framework. Reflecting the culture's hierarchical mind-set, publications are ordered in terms of the outlet's prestige and importance. Journals and publishing houses are each graded by government agencies and accorded a particular status, which then is used to decide the size of the cash prize. For example, if Zhong had one paper published in the US-based journal Science or the UK-based Nature, the two journals considered most prestigious in China, he would have received 100,000 yuan.
The carrot approach seems to have worked well for Jinggangshan University, which was constituted as a teaching college in 2003 after combining a few vocational and technical colleges similar to community colleges in the United States. In 2007, Jinggangshan was granted university status by the Ministry of Education. It offers degrees predominantly at the undergraduate level, with only eighteen master's students in a student body of sixteen thousand. Most of its faculty members, more than one thousand, do not have the necessary training in research, nor are their primary responsibilities researching and publishing. In 2013, about 15 percent of the faculty held a doctoral degree; about one-third had only a bachelor's degree. Yet in the five years before 2003, Jinggangshan University had 4,600 research papers and 254 books published, an achievement as astonishing as the nation's.17
A Chinese Heart
Another force driving China's miracle is patriotism. To inspire scientific and technological innovations, the Chinese government makes good use of the millennium-old Chinese spirit instilled in its intellectuals: love for country.
On February 26, 2003, celebrities and high-level government officials gathered in Shanghai to hear an important announcement. With much fanfare, a panel of leading computer scientists announced their unanimous approval of Hanxin #1, a computer chip invented in China. The experts believed that the chip was the first in China and cutting-edge internationally. Widely reported and celebrated, the announcement marked an important milestone in China's history of chip development.
Three years later, another announcement came. In 2006, after two months of investigation, a panel of experts commissioned by the Ministry of Education, the Ministry of Science and Technology, and the Shanghai government unanimously concluded that “in the research and development process of the Hanxin series of chips, Chen Jin has engaged in serious frauds and deception. He deceived the evaluation experts, Shanghai Jiaotong University, R&D teams, local and the central governments, as well as the media and the public.”18
In three years, Chen Jin had gone from national hero to con artist. Who was he?
Chen earned a doctoral degree in computer engineering from the University of Texas in 1997. He then worked at Motorola for about three years before joining the faculty of Shanghai Jiaotong University, one of the most prestigious engineering schools in China (and the alma mater of China's former president, Jiang Zemin). At the university, Chen was recruited to work on computer chips. In 2000, the Chinese government announced that it wanted its “research and production capacity for software to reach or approximate advanced international standards.” By 2010, China was to “become one of the primary research and production bases for the microchip industry in the world.”19 The announcement came with a series of policies and grants propelling the country toward its goal: inventing its own computer technology.
Chen was given funds, personnel, and facility to create the Hanxin Lab in March 2002. The name was carefully chosen to reflect national pride and patriotism. Han means the Chinese people, and xin means chip but also has the same pronunciation as “heart” in Mandarin. Within less than a year, Chen and his small team—fewer than ten, mostly graduate students at Shanghai Jiaotong—presented the miraculous Hanxin #1, a high-end advanced digital signal processing chip that passed the inspection of leading computer experts and won national recognition. By contrast, it had taken Motorola more than three years and team of over one hundred professional engineers to develop one of its chips, the DSP56800E.
Chen's success won him considerable recognition. He was appointed dean of the College of Microelectronics at Shanghai Jiaotong University, awarded the distinguished and exclusive title of Changjiang Scholar by the Ministry of Edu
cation, and named a National Advanced Worker in Science and Technology. Between 2003 and 2005, he received government funding in excess of 100 million yuan (about $18 million) for further R&D. He continued to produce more advanced chips, from Hanxin #2 to Hanxin #5, and filed twelve patent applications for his inventions.
In January 2006, a message accusing Chen of fraud and deception was posted on an online bulletin board. Five months later, it became clear that Chen had never invented the chip. He had bought five Motorola chips and hired migrant workers to carefully replace the Motorola logo with Hanxin's. The “Chinese hearts” were nothing more than commercially available chips manufactured by Motorola.
More Inventions Than Young Edison
Another incentive for Chinese achievement is the lure of extra points for college admissions. Just the rumor that a national patent can boost a high school graduate's chances in the fiercely competitive college admission process was enough to generate hundreds of patents in an ordinary school in Dalian.20
A 2008 story in the Bandao Chenbao (Peninsula Morning News) reported that one ninth-grade class had been granted over twenty patents. Students in this Dalian school had been granted over five hundred patents in the previous three years. “Over 30% of our students have national patents,” a teacher told the reporter, “but ours are not the highest. Of all middle schools in Dalian, our school is only slightly above average in terms of percentage of students owning national patents.”
The phenomenon stretches nationwide. A secondary school in Wuhan collected more than two thousand patents in about eight years. One of its students was granted twelve patents in a single year. The total number of patents granted to elementary and secondary school students increased from two in 2002 to nearly six thousand in 2011, a five-hundred-fold jump in less than ten years. In the seventeen years before 2002, only thirty patents total had been filed by students K–12.21
In students, patent fever is largely driven by real or rumored policies that patents can add extra points to the college admission exam (gaokao). China has decentralized admissions decisions about bonus points based on ethnicity, special talents, and other achievements, allowing provincial government officials and some selected universities to make the judgment calls. Although there is no uniform policy across the nation about whether—or to what degree—a patent can be counted toward college admission scores, which are typically based on gaokao, there is widespread belief that patents can help dramatically. And that belief is deliberately perpetuated by businesses that, for a fee, help students file patent applications.
Inventions certainly helped Yingying Wu, a controversial student at Beijing Normal University. In December 2006, the university held a press conference about her accomplishments. In this highly unusual event, reporters were told that Wu, a twenty-one-year-old senior studying psychology, had made one hundred inventions, held three patents, had been appointed chair member of the Association of Machinery Computing, and had been hired as vice president of Topcoder, an American company that organizes computer programming contests. Wu's inventions and patents had been a critical factor in her admission to the university three years earlier since her gaokao scores were below the university's cutoff score. Her story had already been sent to many media outlets; this particular press conference was to announce her candidacy for China's national Student of the Year Award.22
As soon as her story was told, Wu became the hottest college student in China. But the initial admiration and praise were quickly replaced by a flurry of online postings and media reports casting doubt on her accomplishments.23 The truth is that her accomplishments were grossly exaggerated. Yes, Topcoder had made her vice president for Asia, but she was the only employee in the Asian office, and Topcoder was a much smaller company than the media were led to believe. She was not a chair member of ACM but an assistant to the one of its committees. She held not three patents but two. There was no evidence of her one hundred inventions, a total that would have outstripped Thomas Edison at the same age. After initially refusing to respond to media inquiries about the accuracy of Wu's story, officials at Beijing Normal University admitted that there were some inaccuracies and exaggeration in Wu's résumé.24 Her stardom ended soon after it began.
Little Cleverness and Junk Papers
Wu and other young inventers have contributed significantly to the total number of China's patents. However, a large proportion of China's patents have been called “junk patents” or, at best, xiao congming (small cleverness). “The corporate and academic opinion is that ‘over 50% or even 80% of Chinese patents are junk,’ ” notes a reporter at China IP Magazine.25 One of Wu's patents, for instance, was called Rapid Search Method and Its Dictionary, which according to the abstract was “a method to quickly find a word in a dictionary through the index attached at the side of the dictionary. The index indicates the location in the dictionary of the first and second syllable of a word.”26 This was essentially the same method that the American Heritage Dictionary series was already using.
Wu's patents belonged to the “utility model” category, which most developed countries do not consider true inventions. Granted, she filed her first patent application at the age of fifteen, an impressive accomplishment for a teenager. But while such patents might prove useful, they are unlikely to be commercialized and bring significant value to the economy. They're not likely to help realize China's dream for true scientific and technological innovations.
Patents like Wu's make up the majority of China's patents. China grants three types of patents: invention, utility model, and design. Invention patents can be granted to both products and processes and must meet standards of novelty, inventiveness, and practical use, determined by a review process called substantive examination. In contrast, utility model patents are granted without review because the expectation of “inventiveness” is far lower. Design patents, like utility model patents, improve products' appearance and are granted without the substantive examination review. Both utility model and design patents are far easier to obtain and far less innovative than invention patents, and they are generally considered to be of lower quality. In many developed countries, invention patents are the only ones granted.
In 2011, out of the 1.5 million patent applications received by the State Intellectual Property Office in China from both domestic and foreign entities in 2011, just about one-third were invention patents. More, 36 percent, were utility model patent applications. Most of the utility applications came from Chinese entities, which filed 1.5 million patent applications all told. Only 28 percent of that total were invention patents; 38 percent were utility model patents. In comparison, 86 percent of patent applications from foreign entities were invention patents, and only 3 percent were utility model patents.
In 2012, the number of patent applications had another significant increase, but the proportion of invention patents remained low. In more than 2 million applications, the percentage of invention patents remained about one-third. The percentage was much lower in the number of patents actually granted—about 217,000 out of 1,255,000, so about 18 percent.27
The dubious quality of patents in China has been openly acknowledged and widely discussed by Chinese officials. “Our nation's patent quality remains at a relatively low level,” said Tian Lipu, director of China's State Office of Intellectual Property, at a press conference in 2013. “Unlike the cutting-edge and innovative high-tech patents, a lot more of our patents are improvement inventions, lower quality utility model and design patents.”28 A comprehensive study of China's patent quality, released in 2012 by the European Union Chamber of Commerce, concluded, “Analysis of a variety of patent statistics suggests that China's progress in patent quality lags behind its rates of patent filings,” as indicated by a broad range of measures:
There are higher ratios of domestic to foreign filings of invention patents in EU countries sampled than in China. There are significantly lower average life-spans of Chinese patents and lower percentages of patents in-force owne
d by domestic filers vs. foreign filers in China compared with the rates in EU and other countries sampled; higher rates of utility model invalidations than invention patent and design patent invalidations; concerning rates of patents filed solely for malicious prosecution actions, which may be made up more so of utility models than other types of patents; poor scores in terms of patent citations; and empirical econometric analyses generally show foreign enterprises at large do not typically file patents on breakthrough inventions in China. China also has lacklustre scores on several other patent quality indicators.29
China's research publications have the same problem, when tested for quality. Nature reported that a 2010 survey commissioned by the Chinese Ministry of Science found that “roughly one-third of more than 6,000 surveyed across six top institutions admitted to plagiarism, falsification or fabrication.”30 In 2008, Helen Zhang, editor of the prestigious Journal of Zhejiang University-Science, pioneered the use of software to spot plagiarism in submissions. “In almost two years, we find about 31 percent of papers with unreasonable copy[ing] and plagiarism,” she told NPR in 2011.31 In another astonishing case, a medical research paper published in 1997 was found to have been plagiarized six times by twenty-five coauthors at sixteen institutions in 2010.32 Despite significant advances in medical technology, the plagiarists continued to report similar results using virtually identical language.
Fraud aside, the overall lower quality of China's research papers is also indicated by another figure: the average number of times a paper has been referenced by other papers. Although the overall number of citations of Chinese papers has increased significantly, the average remains much lower than the world's average: 6.92 times compared to 10.69 times.33