The modern scientific enterprise rests on a deceptively simple idea: knowledge grows when it is shared. Scientific discoveries achieve their greatest value not when they remain confined to laboratories or institutions, but when they are communicated, scrutinized, challenged and extended by others. For more than three centuries, scholarly publishing has provided the infrastructure that enables this collective process, transforming individual observations into humanity's cumulative knowledge.

This was the original purpose of scientific publishing. Journals were created to preserve discoveries, establish priority, encourage independent verification and facilitate scientific dialogue. Publication was a means to an end, the advancement of knowledge, not an end in itself.

Today, however, the landscape of scientific publishing has changed dramatically. Publications are no longer viewed solely as records of discovery. They have become indicators of academic productivity, institutional reputation and professional success. Researchers are recruited, promoted and funded on the basis of publication counts, citation indices, journal rankings and impact factors. Universities compete through research metrics, while governments increasingly rely on bibliometric indicators to evaluate national research performance (Hicks et al., 2015; Nature Editorial, 2022).

The use of metrics is not inherently problematic. Science requires transparent and objective methods to evaluate research performance, allocate resources and monitor progress. Citation analysis, journal metrics and bibliometric indicators can provide useful information when interpreted carefully and within context. Difficulties arise when these indicators become substitutes for informed scientific judgement rather than tools that support it. The distinction is crucial.

The question facing researchers has gradually shifted from "What is the most important scientific problem?" to "What is most likely to be published?" Although subtle, this change has profound implications for how science is conducted. Long-term investigations with uncertain outcomes become less attractive than projects capable of generating multiple publications. Replication studies, negative findings and locally relevant research often struggle for recognition because they attract fewer citations or are perceived as less prestigious. Scientific curiosity increasingly competes with publication incentives.

This phenomenon is often described as the "publish or perish" culture. Yet the phrase understates the complexity of the problem. The challenge is not that researchers publish too much. On the contrary, the remarkable expansion of global scientific literature reflects unprecedented intellectual productivity. The deeper concern is that publication has increasingly become a performance indicator rather than a communication process.

This transformation illustrates a broader principle in public policy and economics known as Goodhart's Law: once a measure becomes the target of decision-making, it gradually loses its effectiveness as a measure. Bibliometric indicators were originally developed to describe patterns of scholarly communication. They were never intended to become definitive measures of scientific excellence. Nevertheless, impact factors, h-indices and citation counts have evolved into benchmarks that influence careers, funding decisions and institutional rankings (Hicks et al., 2015).

The consequences are becoming increasingly apparent. Around the world, concerns have emerged regarding excessive self-citation, citation cartels, honorary authorship and other practices that artificially inflate bibliometric indicators. A recent investigation reported by The Scientist, based on work by Retraction Watch, described an alleged international citation network in which unusually rapid growth in citation counts and h-index values prompted scrutiny by research-integrity experts (Borrell, 2026). Whether or not every allegation is ultimately substantiated, the broader lesson is clear: numerical indicators can be manipulated in ways that do not necessarily reflect genuine scientific influence.

Similarly, the rapid growth of predatory journals demonstrates how incentive systems shape academic behaviour. These journals exploit publication pressures by offering rapid acceptance with little or no meaningful peer review. Researchers, particularly those facing mandatory publication requirements or institutional pressure, may become vulnerable to such outlets despite recognizing their limited scholarly credibility. The problem therefore lies not only with predatory publishers but also with evaluation systems that inadvertently create demand for them.

Even legitimate scholarly publishing faces new tensions. Open-access publishing has transformed the accessibility of research, enabling readers across the world to access scientific knowledge without subscription barriers. At the same time, the widespread adoption of article processing charges (APCs) has shifted financial responsibility from readers to authors. Although this model has expanded access to scientific literature, it has also raised concerns about equity, particularly for researchers working in low-resource institutions and developing countries. A publishing system intended to democratize knowledge should not inadvertently limit who can contribute to it.

Recognizing these challenges, the international research community has begun questioning long-established approaches to research assessment. The San Francisco Declaration on Research Assessment (DORA) argues that journal-based metrics should not be used as surrogate measures of the quality of individual research articles or researchers (American Society for Cell Biology, 2012). Likewise, the Leiden Manifesto emphasizes that quantitative indicators should support, not replace, expert qualitative assessment (Hicks et al., 2015). More recently, Nature has called for research evaluation systems that recognize integrity, openness, collaboration and diverse forms of scholarly contribution rather than relying predominantly on publication-based indicators (Nature Editorial, 2022).

These developments suggest that scientific publishing has reached an important moment of reflection. The challenge is no longer simply how to publish more research, but how to ensure that publishing continues to serve its original purpose.

That purpose deserves to be rediscovered. Rediscovering that purpose requires more than reforming impact factors or tightening editorial policies. It requires changing the narrative that defines academic success.

The central question should no longer be "How many papers did a researcher publish?" but "What lasting contribution did the research make?"

This shift may appear subtle, but it fundamentally changes how scientific achievement is understood.

I propose that scholarly communication should evolve from a Research Publication Ecosystem to a Research Contribution Ecosystem.

The distinction is important. A publication ecosystem values measurable outputs, papers, citations, journal rankings and bibliometric indices. These indicators are relatively easy to count, compare and incorporate into institutional evaluation systems. However, scientific progress has never depended solely on what can be counted.

A contribution ecosystem recognizes that research generates value in many different ways. An openly shared dataset may accelerate hundreds of future studies. A software tool may transform an entire discipline. A replication study may prevent years of scientific error. A policy report may influence national legislation. A technological innovation developed for a remote community may improve thousands of lives despite receiving relatively few academic citations. None of these contributions should be regarded as secondary simply because they produce fewer conventional publications.

Encouragingly, the international research community has already begun moving in this direction. DORA calls for evaluating research on its intrinsic quality rather than the prestige of the publishing journal (American Society for Cell Biology, 2012). The Leiden Manifesto similarly recommends combining quantitative indicators with expert qualitative assessment while recognizing disciplinary diversity and local research priorities (Hicks et al., 2015). More recently, the Coalition for Advancing Research Assessment (CoARA) has urged universities, funding agencies and research organisations to reform evaluation systems so that they better recognise openness, collaboration and broader scientific contributions (CoARA, 2022).

India's recent reforms to doctoral education provide an instructive example of this changing philosophy. The decision to remove the mandatory requirement that PhD candidates publish research papers before thesis submission was interpreted by some as lowering academic standards. A more constructive interpretation is that it restores responsibility to universities and examiners to judge the quality of research directly rather than through publication as a proxy (Vaidyanathan, 2019).

Publication should certainly remain an aspiration of every doctoral scholar. Research that contributes meaningfully to knowledge deserves to be disseminated through peer-reviewed journals. Yet publication should emerge from the quality of research, not from regulatory compulsion.

The same principle extends beyond publication requirements to the broader governance of doctoral education. Many universities continue to require candidates to submit several printed copies of doctoral theses, often running to 400 or 500 pages each. In an era of secure institutional repositories, encrypted digital submissions and electronic examination systems, such practices deserve reconsideration. Maintaining a single archival hard copy, while providing authenticated digital copies for examiners and university libraries, could substantially reduce paper consumption, printing costs and storage requirements. Universities that champion sustainability in research should also demonstrate sustainability in research administration.

Publishers likewise have an important responsibility. Commercial publishers have played a crucial role in disseminating scientific knowledge globally, investing in editorial quality, digital infrastructure and archiving. Nevertheless, the economics of scholarly publishing continue to evolve. Open access has dramatically expanded readership, yet increasing article-processing charges create genuine concerns regarding equity for researchers working in resource-constrained institutions. Ensuring that open science remains inclusive will require continued innovation in publishing models that balance accessibility, financial sustainability and research quality.

Universities, funding agencies and governments must also reconsider how excellence is recognised. Promotion systems built predominantly around publication numbers inevitably reward publication behaviour. If institutions wish to encourage interdisciplinary collaboration, reproducible research, public engagement, mentorship, innovation and societal impact, then these activities must become visible within evaluation frameworks. Researchers respond to incentives. Changing evaluation systems therefore changes research culture.

This is not an argument against bibliometric indicators. Impact factors, citation analyses and research metrics remain valuable tools for understanding scientific communication. The mistake lies in expecting any single metric to capture the complexity of scientific achievement. As science itself becomes increasingly interdisciplinary, collaborative and socially engaged, research assessment must become equally sophisticated.

Ultimately, the future of scientific publishing depends less on technology than on philosophy.

The digital revolution has transformed how science is communicated. Artificial intelligence is transforming how knowledge is generated, reviewed and synthesised. Open science is reshaping access to information. Yet none of these innovations can answer the most fundamental question facing scientific publishing:

Why do scientists publish?

If the answer is to satisfy institutional metrics, improve rankings or accumulate citations, then scientific publishing risks becoming an administrative exercise rather than an intellectual enterprise.

If the answer is to communicate reliable knowledge, invite critical scrutiny and improve the human condition, then the path forward becomes clearer.

Scientific publishing should once again become what it was originally intended to be: a trusted public record of humanity's search for truth.

The greatest legacy of science has never been the number of papers it produced. Its legacy has been the knowledge it created. Perhaps the future of scientific publishing should be judged in exactly the same way.

References 

American Society for Cell Biology. (2012). San Francisco Declaration on Research Assessment (DORA). https://sfdora.org

Borrell, B. (2026). A researcher's suspiciously high h-index revealed a vast citation ring. The Scientist. https://www.the-scientist.com/a-researcher-s-suspiciously-high-h-index-revealed-a-vast-citation-ring-74625

Coalition for Advancing Research Assessment. (2022). Agreement on reforming research assessment. https://coara.eu

Hicks, D., Wouters, P., Waltman, L., de Rijcke, S., & Rafols, I. (2015). Bibliometrics: The Leiden Manifesto for research metrics. Nature, 520(7548), 429–431. https://doi.org/10.1038/520429a

Nature Editorial. (2022). Research evaluation needs to change with the times. Nature, 601(7894), 166. https://doi.org/10.1038/d41586-022-00056-z

OECD. (2025). OECD Science, Technology and Innovation Outlook 2025: Driving Change in a Shifting Landscape. OECD Publishing.

STM Association. (2024). STM Global Brief 2024.

UNESCO. (2021). UNESCO Science Report: The Race Against Time for Smarter Development. UNESCO Publishing.

Vaidyanathan, G. (2019). No paper, no PhD? India rethinks graduate student policy. Nature. https://doi.org/10.1038/d41586-019-01692-8

Clarivate. (2025). Journal Citation Reports. Clarivate.