Failure is the most likely option

Reject. Reject. Reject. Reject & Resubmit, then Accept. Major Revision. Does this look like the record of a successful scientist? Indeed, this is the very brief history of the five scientific articles I submitted to Materials Horizons, between 2023 and 2026. You may think that this is not something one should be proud of or disclose publicly.

My name is Carlo Antonini; I am an aerospace engineer with a PhD in industrial engineering, who began a tenure-track position at the Department of Materials Science at the University of Milano-Bicocca, Italy, in 2018 and was (successfully) tenured in 2021. If you can spare a few minutes to read this Editorial, I will try to explain why I think it is important to speak about failure – at least, the submission record disclosed above should qualify me to do so.

Some of you may be familiar with “A CV of failures”, written by Melanie Stefan in 2010.¹ Since then, several colleagues have started tracking their major failures (Johannes Haushofer provides a particularly good example).² When I first read about it, a few years ago, I immediately thought I should write and publish mine; however, I was initially concerned that publishing such a record could be misperceived as a complaint or a whine. Or, more simply, I was afraid to show that I was not successful enough.

In April 2022, I finally decided it was time to compile my own “CV of failures”³ and published a post on LinkedIn with an explanation of my decision.⁴

Why we need to talk about failure. As professionals and scientists, and more broadly as humans living in the age of social media, we feel the need to continuously brand ourselves to secure a new job, a promotion, or a grant. We clearly put more effort than previous generations into crafting and promoting our personal narratives. This is not necessarily negative: for example, we now have the chance to directly communicate the beauty of our science, discoveries and inventions to a larger audience. Before social media, this was a much longer, more heavily filtered process. However, social media have their downsides. On these platforms, it is common to find people celebrating only successes, like a new job, a published paper or a granted proposal. It is hard to let “real life” – made of failures, repetitive tasks, and the occasionally boring work required to ensure reproducibility – emerge. This is well captured in the movie “Joy, the story of IVF” (In Vitro Fertilization), based on the true story of the three pioneering British scientists and their struggles to develop IVF. Dr Robert Edwards, the biologist in the team, tells his colleague: “I am a scientist, Jean. I fail every day, in the hope that I’ll succeed in twenty years.”

Defining success and failure. Using a scientific approach, we should first start defining success and failure and how to measure them. According to the Oxford dictionary, success is “the achievement of a desired result or outcome; the accomplishment of an aim or purpose”, while failure is “the fact of failing to effect one's purpose”. In the past, success was often associated strictly with one's professional career; however, among Millennials – to whom I belong – and even more for GenZs, the definition has broadened. There is an overall shift towards a more heuristic view of success, for example in the balance between work and life. Nonetheless, I will restrict the discussion here to what it means to be successful as a scientist, specifically within materials science.

Why is failure so common and pervasive? Science is a human endeavour and, as such, it evolves through a process akin natural selection, which is characterised by a long sequence of mistakes and failures. Just as nature selects certain forms of life, ideas in science must also undergo selection. Because resources are limited, we must focus on only a few of them. This can be a painful process – for example, only 10% of proposals may be granted in the most competitive calls – but this is also how ideas get refined, improved, and become feasible. If we had endless resources and a success rate close to 100%, we would be tempted to write as many proposals as possible, but we would not necessarily invest enough thought into them. I am obviously not suggesting that it is positive to have limited resources for science, but I want to stress that hard selection processes allow great ideas to emerge. Furthermore, natural selection teaches us that “It is not the strongest of the species that survives, nor the most intelligent that survives. It is the one that is the most adaptable to change” (a quote often mistakenly attributed to Darwin, but actually spoken by Leon C. Megginson regarding Darwin's theories). This means that as scientists, we need to understand that many things beyond our control may condition our careers: changes in funding priorities, personal needs (e.g. caretaking), visa restrictions, or even wars, may affect our professional journey. We must be able to adapt to those conditions to grow as professionals. As an aerospace engineer, my shift towards materials science over the years was driven not only by a genuine interest that started to emerge during my PhD and early post-doc, but was certainly facilitated by the great boom of the materials science in terms of job and funding opportunities in the last two decades.

How should we ultimately measure success? A few weeks ago, I attended the ACIS2026 conference, where Kaye Minkyung Kang, a senior lecturer at the University of Sydney, gave an invited talk including a discussion on her career. She mentioned that as a scientist, you can always find someone telling you that what you are doing is not good enough. I can add that sometimes that person may be yourself: this is called imposter syndrome. However, the speaker shared that each of us can still define what success means for ourselves. I am sure she did not mean that success is entirely subjective, but rather that in science there is no pre-packaged checklist to becoming successful. Also, moments of genuine criticism – that may feel like a failure – can become an opportunity to improve your ideas and bring them to the next level.

We have a major issue in science: unlike a company having the explicit objective of generating profit, a research group aims to generate knowledge, which is very hard to measure objectively. As a result, proxies are used: just as a company is assessed by market capitalization or price-to-earnings ratio, academia uses quantitative inputs such as papers and citations. It is correct to identify quantitative measures, as long as papers and citations do not become the final goal and only metrics of research. Nonetheless, we may end up confusing the ultimate goal of research with proxies, because when faced with a difficult question (the target question), the human mind, and more precisely the so-called system 1 – the fast, intuitive part of your brain – searches for a related question that is easier to answer (the heuristic question) and offers an answer to that question, instead.⁵

A career in science is an infinite game. We apply the same approach when assessing someone's (or our own) careers. To understand this, it is helpful to look at the concepts of finite and infinite games, introduced by philosopher James P. Carse in the 80s and popularized by Simon Sinek. There are at least two kinds of games: finite and infinite (Fig. 1). A finite game is played for the purpose of winning, an infinite game for the purpose of continuing the play. Finite games are those instrumental activities – from sports to politics to wars – in which the participants obey rules, recognize boundaries and announce winners and losers. The infinite game includes any authentic interaction, where rules and boundaries change, and we play it solely for the purpose of continuing the game. We often view a career as a finite game, trying to balance victories (successes) against losses (failures). This way of thinking comes naturally because a career is made of milestones: you pass or fail an exam, a proposal is granted or not, a paper is accepted or rejected.

Fig. 1 Schematic comparison of a finite vs. infinite games. Image generated with AI using NotebookLM.

However, if you see your career as a finite game, everyone else becomes a competitor you need to beat. We need to a shift towards an infinite game mindset. In this view, failure is never definitive: a “loss”, like a rejected paper or a missed opportunity, is not a game over screen, but a temporary dip in the narrative. Also, there is no ultimate “winner”: as Darwinian evolution teaches us, external conditions and rules may change. The goal is resilience: you need to be capable to adapt and continue to play.

As a professor, my primary duties are teaching, research, and service. In a maybe extreme but still possible future scenario, AI may become so pervasive that fundamental courses will be completely AI-based, and I could eventually not be able to cope with a new way of doing research. Will I be fired? Will I still be “successful”, as I may be considered now because of my tenure as faculty?

To navigate through uncertainty, I keep a few guidelines in mind, and I would like to share them with you.

Never stop learning and improving. Adopt a growth mindset. Aim to be better today than you were yesterday to ensure you can keep playing tomorrow. Ask yourself: what I have learnt today? Or what would I learn if I were involved in that project or initiative? We have a unique privilege in science, which is continuing to learn. Take advantage of it, and enjoy it, too.

Invest in trust and cooperation with others. You will be playing with the same people for a long time; building bridges is more effective than burning them for a quick win. There are many occasions where I get impressed by the help I receive when I ask, even from people I would not expect. Make sure you have a close circle of trusted people, and also you take advantage of “The Strength of Weak Ties”⁶ through an extended network of people that can help you when you need it.

Answer a simple deep question: what is my contribution? Instead of asking yourself what benefit is in it for you, choose projects that align with your core values – whether that is addressing climate change, improving health, or developing STEM education programs. Once you have identified such values, it will be more natural for you to study and work hard on something that aligns with them. This was beautifully summarized by Ben Horowitz to the class of 2015 at Columbia University: “Don’t follow your passion, follow your contribution”.⁷ I find that this advice offers a pragmatic approach to reconcile the idealistic just-do-what-you-like, with the realistic need that we need to live on our salary. Contributing to something promotes internal motivation, makes us grow and learn, and is one of the most effective drivers for improvement.

Short- and long-term objectives. A journey in one career is like driving a car: you need to pay attention to your long-term destination, while remaining aware of immediate surroundings. In the short term, keep track of goals to stay focused. As a supervisor, I have yearly discussions with the PhDs and post-docs in the group to define priorities on a scale from 1 (necessary) to 3 (great to have). After a year, I ask the students to self-assess their achievements. Plans may change during the year, but the list can help in two ways: (i) when one feels overwhelmed, it provides a clear direction; (ii) it is rewarding to see that important milestones were reached. At the same time, keep track of long-term scenarios. The numbers in academia are stark: by simple back of the envelope calculation, a professor may supervise 10 to 20 PhD students, but only one is needed to replace them in a steady state. This means that the chance of becoming a professor after a PhD is roughly 5-10%. And even if you make it and become a professor, many relevant grants (e.g., fundamental science competitive EU grants) have similar success rates. You’ll continue to struggle to acquire resources to do great science. As a scientist, you are well trained to think in terms of scenarios: use those skills to navigate uncertainties.

Learn to frame and tell your story. Learn to frame your story through a narrative CV, which goes beyond a list of achievements, to make your scientific ambitions and growth process explicit. The first time I had to prepare one was for an ERC (Starting) Grant, arguably the most prestigious individual research grant in the EU, and it gave me the opportunity to self-reflect on my long-term scientific growth and ambitions (note for the reader: my proposal was not granted). Within the EU, the Marie Curie Alumni Association, an association gathering current and previous recipients of Marie Sklodowska Curie Actions, recently promoted narrative CV as new complementary information to be provided in a grant application.⁸

To conclude, I hope I have helped you to look at failure from a different perspective. Perhaps you will start keeping track of your own failures to see how they helped you to get where you are – or even write your own “CV of failures”. By the way, some of you may wonder if there is a newer version of my own “CV of failures” after the 2022 edition. I promised to myself that I would update it every couple of years, but I have to confess that, so far, I have failed to keep that promise.

Acknowledgements

I would like to acknowledge all colleagues who have failed with me during the journey, so far. More failures will come! I acknowledge Aldo Ferrari, former ETH Zurich colleague and currently CSO at the company Hylomorph, for the long discussions on science and life in academia. I also acknowledge Alidad Amirfazli, PhD co-supervisor and currently professor at York University, for the “car driving” metaphor (and much more than that). I also acknowledge Gene Kranz, NASA Flight Director and the Apollo 13 Moon landing mission, for the quote “Failure is not an option”, which clearly inspired the title of this editorial. Funnily enough, that mission was technically a failure, as the astronauts did not land on the Moon.

Notebook LM was used to generate Fig. 1, and Gemini was used to check grammar and improve text readability. The author is fully responsible and accountable for the content of their article, including any parts produced by an AI tool.

References

1. M. Stefan, A CV of failures, Nature, 2010, 468, 467,  DOI:10.1038/nj7322-467a.
2. https://johanneshaushofer.com/cv-bio.
3. https://sefilab.mater.unimib.it/carlo-antonini.
4. https://www.linkedin.com/pulse/what-failure-just-most-likely-option-carlo-antonini/.
5. D. Kahneman, Thinking, fast and slow, 2011.
6. M. S. Granovetter, The Strength of Weak Ties, The American Journal of Sociology, 1973, vol. 78, No. 6, pp. 1360–1380.
7. https://www.youtube.com/watch?v=uaSqh4DiQSw.
8. https://medium.com/marie-curie-alumni/narrative-cv-the-what-how-and-why-1e7525f17424.

---