Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Seismology: Japan must admit it can't predict quakes

Japan's government is still failing to protect the public against earthquakes, six years on from the devastating magnitude-9.1 Tohoku earthquake of 2011. In my view, it urgently needs to abandon flawed, old policies and focus instead on implementing effective countermeasures that are based on sound research.

Despite my earlier advice (R. J. Geller Nature 472, 407–409; 2011), the government continues to claim that a magnitude-9 earthquake is highly likely in the Tokai–Nankai district during the next 30 years. This claim is based on the false hypothesis that great earthquakes of magnitude 8 or more occur cyclically (see Y. Y. Kagan et al. Seismol. Res. Lett. 83, 951–953; 2012). Clinging to such repudiated models even after the 'unexpected' 2011 Tohoku quake is, in my view, cognitive dissonance on a massive scale.

In my 2011 Comment article, I also advocated repeal of Japan's 1978 Large-Scale Earthquake Countermeasures Act (LECA). LECA is based on the false premise that precursors of a great earthquake in the Tokai district can be detected accurately enough for a legally binding three-day state of emergency to be declared. However, the act is still in place. The government intends to tweak it to cover a larger land area, while authorizing the issuance of warnings that fall short of a state of emergency (see 'Don't rely on quake predictions', The Japan Times 7 July 2016).

I urge the government to explicitly inform the public that prediction of imminent earthquakes is currently impossible. It should repeal LECA and adopt earthquake countermeasures that are based on sound science.

Author information



Corresponding author

Correspondence to Robert J. Geller.

Related links

Related links

Related links in Nature Research

Shake-up time for Japanese seismology

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Geller, R. Seismology: Japan must admit it can't predict quakes. Nature 545, 289 (2017).

Download citation


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing