To Engineer Is Human: The Role of Failure in Successful Design

To Engineer Is Human: The Role of Failure in Successful Design

by Henry Petroski


$15.26 $16.95 Save 10% Current price is $15.26, Original price is $16.95. You Save 10%. View All Available Formats & Editions
Choose Expedited Shipping at checkout for guaranteed delivery by Monday, December 16


How did a simple design error cause one of the great disasters of the 1980s - the collapse of the walkways at the Kansas City Hyatt Regency Hotel? What made the graceful and innovative Tacoma Narrows Bridge twist apart in a mild wind in 1940? How did an oversized waterlily inspire the magnificent Crystal Palace, the crowning achievement of Victorian architecture and engineering? These are some of the failures and successes that Henry Petroski, author of the acclaimed The Pencil, examines in this engaging, wonderfully literate book. More than a series of fascinating case studies, To Engineer is Human is a work that looks at our deepest notions of progress and perfection, tracing the fine connection between the quantifiable realm of science and the chaotic realities of everyday life.

Product Details

ISBN-13: 9780679734161
Publisher: Knopf Doubleday Publishing Group
Publication date: 03/28/1992
Edition description: Reprint
Pages: 272
Sales rank: 67,745
Product dimensions: 5.19(w) x 8.00(h) x 0.58(d)

About the Author

Henry Petroski is the Aleksandar S. Vesic Professor of Civil Engineering and a professor of history at Duke University. The author of more than a dozen previous books, he lives in Durham, North Carolina, and Arrowsic, Maine.

Read an Excerpt



Shortly after the Kansas City Hyatt Regency Hotel skywalks collapsed in 1981, one of my neighbors asked me how such a thing could happen. He wondered, did engineers not even know enough to build so simple a structure as an elevated walkway? He also recited to me the Tacoma Narrows Bridge collapse, the American Airlines DC–10 crash in Chicago, and other famous failures, throwing in a few things he had heard about hypothetical nuclear power plant accidents that were sure to exceed Three Mile Island in radiation release, as if to present an open-andshut case that engineers did not quite have the world of their making under control.

I told my neighbor that predicting the strength and behavior of engineering structures is not always so simple and well-defined an undertaking as it might at first seem, but I do not think that I changed his mind about anything with my abstract generalizations and vague apologies. As I left him tending his vegetable garden and continued my walk toward home, I admitted to myself that I had not answered his question because I had not conveyed to him what engineering is. Without doing that I could not hope to explain what could go wrong with the products of engineering. In the years since the Hyatt Regency disaster I have thought a great deal about how I might explain the next technological embarrassment to an inquiring layman, and I have looked for examples not in the esoteric but in the commonplace. But I have also learned that collections of examples, no matter how vivid, no more make an explanation than do piles of beams and girders make a bridge.

Engineering has as its principal object not the given world but the world that engineers themselves create. And that world does not have the constancy of a honeycomb's design, changeless through countless generations of honeybees, for human structures involve constant and rapid evolution. It is not simply that we like change for the sake of change, though some may say that is reason enough. It is that human tastes, resources, and ambitions do not stay constant. We humans like our structures to be as fashionable as our art; we like extravagance when we are well off, and we grudgingly economize when times are not so good. And we like bigger, taller, longer things in ways that honeybees do not or cannot. All of these extra-engineering considerations make the task of the engineer perhaps more exciting and certainly less routine than that of an insect. But this constant change also introduces many more aspects to the design and analysis of engineering structures than there are in the structures of unimproved nature, and constant change means that there are many more ways in which something can go wrong.

Engineering is a human endeavor and thus it is subject to error. Some engineering errors are merely annoying, as when a new concrete building develops cracks that blemish it as it settles; some errors seem humanly unforgivable, as when a bridge collapses and causes the death of those who had taken its soundness for granted. Each age has had its share of technological annoyances and structural disasters, and one would think engineers might have learned by now from their mistakes how to avoid them. But recent years have seen some of the most costly structural accidents in terms of human life, misery, and anxiety, so that the record presents a confusing image of technological advancement that may cause some to ask, "Where is our progress?"

Any popular list of technological horror stories usually comprises the latest examples of accidents, failures, and flawed products. This catalog changes constantly as new disasters displace the old, but almost any list is representative of how varied the list itself can be. In 1979, when accidents seemed to be occurring left and right, anyone could rattle off a number of technological embarrassments that were fresh in everyone's mind, and there was no need to refer to old examples like the Tacoma Narrows Bridge to make the point. It seemed technology was running amok, and editorial pages across the country were anticipating the damage that might occur as the orbiting eighty-five-ton Skylab made its unplanned reentry. Many of the same newspapers also carried the cartoonist Tony Auth's solution to the problem. His cartoon shows the falling Skylab striking a flying DC–10, itself loaded with Ford Pintos fitted with Firestone 500 tires, with the entire wreckage falling on Three Mile Island, where the fire would be extinguished with asbestos hair dryers.

While such a variety may be unique to our times, the failure of the products of engineering is not. Almost four thousand years ago a number of Babylonian legal decisions were collected in what has come to be known as the Code of Hammurabi, after the sixth ruler of the First Dynasty of Babylon. There among nearly three hundred ancient cuneiform inscriptions governing matters like the status of women and drinking-house regulations are several that relate directly to the construction of dwellings and the responsibility for their safety:

If a builder build a house for a man and do not make its construction firm, and the house which he has built collapse and cause the death of the owner of the house, that builder shall be put to death.

If it cause the death of the son of the owner of the house, they shall put to death a son of that builder.

If it cause the death of a slave of the owner of the house, he shall give to the owner of the house a slave of equal value.

If it destroy property, he shall restore whatever it destroyed, and because he did not make the house which he built firm and it collapsed, he shall rebuild the house which collapsed from his own property.

If a builder build a house for a man and do not make its construction meet the requirements and a wall fall in, that builder shall strengthen the wall at his own expense.

This is a far cry from what happened in the wake of the collapse of the Hyatt Regency walkways, subsequently found to be far weaker than the Kansas City Building Code required. Amid a tangle of expert opinions, $3 billion in lawsuits were filed in the months after the collapse of the skywalks. Persons in the hotel the night of the accident were later offered $1,000 to sign on the dotted line, waiving all subsequent claims against the builder, the hotel, or anyone else they might have sued. And today opinions as to guilt or innocence in the Hyatt accident remain far from unanimous. After twenty months of investigation, the U. S. attorney and the Jackson County, Missouri, prosecutor jointly announced that they had found no evidence that a crime had been committed in connection with the accident. The attorney general of Missouri saw it differently, however, and he charged the engineers with "gross negligence." The engineers involved stand to lose their professional licenses but not their lives, but the verdict is still not in as I write three years after the accident.

The Kansas City tragedy was front-page news because it represented the largest loss of life from a building collapse in the history of the United States. The fact that it was news attests to the fact that countless buildings and structures, many with designs no less unique or daring than that of the hotel, are unremarkably safe. Estimates of the probability that a particular reinforced concrete or steel building in a technologically advanced country like the United States or England will fail in a given year range from one in a million to one in a hundred trillion, and the probability of death from a structural failure is approximately one in ten million per year. This is equivalent to a total of about twenty-five deaths per year in the United States, so that 114 persons killed in one accident in Kansas City was indeed news.

Automobile accidents claim on the order of fifty thousand American lives per year, but so many of these fatalities occur one or two at a time that they fail to create a sensational impact on the public. It seems to be only over holiday weekends, when the cumulative number of individual auto deaths reaches into the hundreds, that we acknowledge the severity of this chronic risk in our society. Otherwise, if an auto accident makes the front page or the evening news it is generally because an unusually large number of people or a person of note is involved. While there may be an exception if the dog is famous, the old saying that "dog bites man" is not news but that "man bites dog" is, applies.

We are both fascinated by and uncomfortable with the unfamiliar. When it was a relatively new technology, many people eschewed air travel for fear of a crash. Even now, when aviation relies on a well-established technology, many adults who do not think twice about the risks of driving an automobile are apprehensive about flying. They tell each other old jokes about white-knuckle air travelers, but younger generations who have come to use the airplane as naturally as their parents used the railroad and the automobile do not get the joke. Theirs is the rational attitude, for air travel is safe, the 1979 DC–10 crash in Chicago notwithstanding. Two years after that accident, the Federal Aviation Administration was able to announce that in the period covering 1980 and 1981, domestic airlines operated without a single fatal accident involving a large passenger jet. During the period of record, over half a billion passengers flew on ten million flights. Experience has proven that the risks of technology are very controllable.

However, as wars make clear, government administrations value their fiscal and political health as well as the lives of their citizens, and sometimes these objectives can be in conflict. The risks that engineered structures pose to human life and environments pose to society often conflict with the risks to the economy that striving for absolute and perfect safety would bring. We all know and daily make the trade-offs between our own lives and our pocketbooks, such as when we drive economy-sized automobiles that are incontrovertibly less safe than heavier-built ones. The introduction of seat belts, impact-absorbing bumpers, and emission-control devices have contributed to reducing risks, but gains like these have been achieved at a price to the consumer. Further improvements will take more time to perfect and will add still more to the price of a car, as the development of the air bag system has demonstrated. Thus there is a constant tension between manufacturers and consumer advocates to produce safe cars at reasonable prices.

So it is with engineering and public safety. All bridges and buildings could be built ten times as strong as they presently are, but at a tremendous increase in cost, whether financed by taxes or private investment. And, it would be argued, why ten times stronger? Since so few bridges and buildings collapse now, surely ten times stronger would be structural overkill. Such ultraconservatism would strain our economy and make our built environment so bulky and massive that architecture and style as we know them would have to undergo radical change. No, it would be argued, ten times is too much stronger. How about five? But five might also arguably be considered too strong, and a haggling over numbers representing no change from the present specifications and those representing five- or a thousand-percent improvement in strength might go on for as long as Zeno imagined it would take him to get from here to there. But less-developed countries may not have the luxury to argue about risk or debate paradoxes, and thus their buildings and boilers can be expected to collapse and explode with what appears to us to be uncommon frequency.

Callous though it may seem, the effects of structural reliability can be measured not only in terms of cost in human lives but also in material terms. This was done in a recent study conducted by the National Bureau of Standards with the assistance of Battelle Columbus Laboratories. The study found that fracture, which included such diverse phenomena as the breaking of eyeglasses, the cracking of highway pavement, the collapse of bridges, and the breakdown of machinery, costs well over $100 billion annually, not only for actual but also for anticipated replacement of broken parts and for structural insurance against parts breaking in the first place. Primarily associated with the transportation and construction industries, many of these expenses arise through the prevention of fracture by overdesign (making things heavier than otherwise necessary) and maintenance (watching for cracks to develop), and through the capital equipment investment costs involved in keeping spare parts on hand in anticipation of failures. The 1983 report further concludes that the costs associated with fracture could be reduced by one half by our better utilizing available technology and by improved techniques of fracture control expected from future research and development.

Recent studies of the condition of our infrastructure — the water supply and sewer systems, and the networks of highways and bridges that we by and large take for granted — conclude that it has been so sorely neglected in many areas of the country that it would take billions upon billions of dollars to put things back in shape. (Some estimates put the total bill as high as $3 trillion.) This condition resulted in part from maintenance being put off to save money during years when energy and personnel costs were taking ever-larger slices of municipal budget pies. Some water pipes in large cities like New York are one hundred or more years old, and they were neither designed nor expected to last forever. Ideally, such pipes should be replaced on an ongoing basis to keep the whole water supply system in a reasonably sound condition, so that sudden water main breaks occur very infrequently. Such breaks can have staggering consequences, as when a main installed in 1915 broke in 1983 in midtown Manhattan and flooded an underground power station, causing a fire. The failure of six transformers interrupted electrical service for several days. These happened to be the same days of the year that ten thousand buyers from across the country visited New York's garment district to purchase the next season's lines. The area covered by the blackout just happened to be the blocks containing the showrooms of the clothing industry, so that there was mayhem where there would ordinarily have been only madness. Financial losses due to disrupted business were put in the millions.

In order to understand how engineers endeavor to insure against such structural, mechanical, and systems failures, and thereby also to understand how mistakes can be made and accidents with far-reaching consequences can occur, it is necessary to understand, at least partly, the nature of engineering design. It is the process of design, in which diverse parts of the "given-world" of the scientist and the "made-world" of the engineer are reformed and assembled into something the likes of which Nature had not dreamed, that divorces engineering from science and marries it to art. While the practice of engineering may involve as much technical experience as the poet brings to the blank page, the painter to the empty canvas, or the composer to the silent keyboard, the understanding and appreciation of the process and products of engineering are no less accessible than a poem, a painting, or a piece of music. Indeed, just as we all have experienced the rudiments of artistic creativity in the childhood masterpieces our parents were so proud of, so we have all experienced the essence of structural engineering in our learning to balance first our bodies and later our blocks in ever more ambitious positions. We have learned to endure the most boring of cocktail parties without the social accident of either our bodies or our glasses succumbing to the force of gravity, having long ago learned to crawl, sit up, and toddle among our tottering towers of blocks. If we could remember those early efforts of ours to raise ourselves up among the towers of legs of our parents and their friends, then we can begin to appreciate the task and the achievements of engineers, whether they be called builders in Babylon or scientists in Los Alamos. For all of their efforts are to one end: to make something stand that has not stood before, to reassemble Nature into something new, and above all to obviate failure in the effort.

Because man is fallible, so are his constructions, however. Thus the history of structural engineering, indeed the history of engineering in general, may be told in its failures as well as in its triumphs. Success may be grand, but disappointment can often teach us more. It is for this reason that hardly a history can be written that does not include the classic blunders, which more often than not signal new beginnings and new triumphs. The Code of Hammurabi may have encouraged sound construction of reproducible dwellings, but it could not have encouraged the evolution of the house, not to mention the skyscraper and the bridge, for what builder would have found incentive in the code to build what he believed to be a better but untried house? This is not to say that engineers should be given license to experiment with abandon, but rather to recognize that human nature appears to want to go beyond the past, in building as in art, and that engineering is a human endeavor.


Excerpted from "To Engineer Is Human"
by .
Copyright © 1992 Henry Petroski.
Excerpted by permission of St. Martin's Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Table of Contents

Title Page,
Copyright Notice,
1. Being Human,
2. Falling Down Is Part of Growing Up,
3. Lessons from play; Lessons from Life,
4. Engineering as Hypothesis,
5. Success Is Foreseeing Failure,
6. Design Is Getting From Here to There,
7. Design as Revision,
8. Accidents Waiting to Happen,
9. Safety in Numbers,
10. When Cracks Become Breakthroughs,
11. Of Bus Frames and Knife Blades,
12. Interlude: The Success Story of the Crystal Palace,
13. The Ups and Downs of Bridges,
14. Forensic Engineering and Engineering Fiction,
15. From Slide Rule to Computer: Forgetting How It Used to Be Done,
16. Connoisseurs of Chaos,
17. The Limits of Design,
Praise for Henry Petroski's To Engineer Is Human,

Customer Reviews

Most Helpful Customer Reviews

See All Customer Reviews

To Engineer Is Human: The Role of Failure in Successful Design 3.6 out of 5 based on 0 ratings. 13 reviews.
jorgecardoso on LibraryThing More than 1 year ago
The last post about "The Evolution of Useful Things" reminded me about another of Petroski's books that I read some years ago: "To Engineer is Human: The Role of Failure in Successful Design". It's also about failure but in a different perspective: how failure makes the engineering activity evolve. Failure has, obviously, a tight relation with engineering. A great part of engineering is making sure that something works within certain bounds of acceptable (ab)use in the most efficient manner possible. In this book Petroski describes some well-known catastrophic failures of engineering structures like the Tacoma Narrows Bridge and the Kansas City Hyatt Regency Hotel walkways and how they came to happen and how they served as example to improve subsequent structures. These two books also remind me that scientific research is also much about finding failure and trying to develop a way to mitigate it....
Newmans2001 on LibraryThing More than 1 year ago
When I found this book I could not wait to read it. And it had some very interesting and enlightening information, but I felt the attempt at making a literary masterpiece out of engineering did not succeed very well. His thoughts as an Engineer should be very logical, creative and systematic, but he got lost in his own thoughts and his concept got very tiring and simplistic. Maybe I should not rate the book as I did not finish it.
ablueidol on LibraryThing More than 1 year ago
Read. well tried to read this last week. Clearly I'm no engineer waiting to astound the world so abandoned it after skimming to find the interesting bits - there weren't any. But just know my bookshelves sighed and committed suicide - lots of broken bricks and gouged wood. We isn't it good to know it was something about breaking points and beam alignment. Oh well back to having a life.
bduguid on LibraryThing More than 1 year ago
Henry Petroski is an author of "popular engineering" books, the cousin to "popular science", which attempt to explain the process of engineering design to a non-specialist audience.This book documents how successful engineering is a process of predicting and preventing failure. Several chapters offer a variety of viewpoints on the philosophy of design: engineering as hypothesis (this building will stand up) which is tested analytically or empirically; design as revision (if we change this bit it will stand up); success as foreseeing failure etc.There are several good angles here, particularly where Petroski likens engineering design to the way in which children learn. For non-engineers, there is also useful material on factors of safety, failure by cracking and other basics.Petroski's use of language is excellent, but as an engineer, I do find a lot of the book disappointing. Non-engineers might come away thinking they know why Tacoma Narrows collapsed, or what fatigue cracking is, but the technical reasons are at best alluded to, never properly explained. Petroski's paper-clip example for fatigue cracking is particularly poor, as it mixes in two generally unrelated issues (brittle failure and plastic strain hardening). For technical matters, "Why buildings fall down" by Levy and Salvadori is far superior, and much better illustrated with simple and easy-to-follow diagrams.Where Petroski succeeds is in the human processes of design engineering, but even here he is somewhat weak. He's good on the philosophy but not the reality - you couldn't read this and get any grasp on how a design engineer actually spends their day, for example.Worth reading, but let down by its fear of the technicalities.
Manith_2000 More than 1 year ago
Was one of the first engineering related books I read, and it is very good. It covers a wide range of topics and links them well to the main point of the book: the importance of failure in engineering. Plus, it describes various engineering marvels to prove this point across; the chapter on Joseph Paxton's Crystal Palace is my personal favourite as it highlights a less-known but amazing story of engineering. Though it is an interesting read, it lacks technical depth than I hoped it would have, and was less informative on certain parts. Nonetheless, it is a book which is great for anyone remotely interested in engineering.
Anonymous More than 1 year ago
When I was preparing to apply to become a chartered member of the Institution of Civil Engineers, a mentor – I think as a warning against complacency – challenged me that I should be reading at least 6 books about engineering a year. From my experience, most engineering books are text books, so I asked him which books he suggested, and he couldn’t think of a single example of a readable book about engineering! Petroski’s books then are rare gems of attempting to span that large distance between the remote island of engineering knowledge and the mainland of public imagination. Reading it as a structural engineer there was plenty that I was already familiar with, but also lots that was new. He draws on examples from both sides of the Atlantic, so having learnt my trade in the UK I found the discussion of iron bridges particularly Othmar Ammann (Quebec Bridge) and the Roeblings (Brooklyn Bridge) fresh and new. While others may be less familiar with Beauvais Cathedral, cracks in Big Ben or Paxton and the Crystal Palace. His opening chapters suggest that the book is for a non-technical audience, but he sometimes lapses into advice for professionals such as lamenting that our drawings are no longer as beautiful as Galileo’s. The vocabulary is probably only accessible for teenagers and up (discussion of monographs, commissions, cantilevers etc.) but other than that, in my opinion it is accessible for a non-technical reader. If I was being picky, I would like... continued at
Subway_Reader More than 1 year ago
The most interesting thing about the book is the cover (a famous photograph of the Tacoma Narrows Bridge collapsing in the wind in 1940). But then all it really amounts to is 232 pages of defensiveness, repeating over and over that failure is just a part of life and experimentation and if we don't want to stagnate we have to expect cracks and collapses, so don't blame the hapless engineer for trying to forward human achievement. Yes, engineers and designers can sometimes be the scapegoats for failures beyond their control, but he makes a weak case and uses this argument to eclipse the real causes and analyses of structural failure. The author cites very few actual examples, analizes the same few examples over and over, and his one or two attempts at discussing novel failures--such as cracks in a set of kitchen knives--go nowhere since it's pure speculation, with no actual research. What's most glaring is his almost complete neglect for the basic role of cost-cutting measures in leading businesses to ignore testing, quality control, maintenance, and basic oversight, putting profits before safety, an essential component he barely touches. In a book supposedly written for the lay person the illustrations, which are critical for understanding the examples he refers to, are few and far between. Some of the major examples he cites, such as the collapse of the Hartford Civic Center roof or the John Hancock building windows failure in Boston, aren't illustrated at all. And for a mostly polemical work there are surprising inconsistencies--in the last chapter he says that that exposing engineering failure are a credit to the profession, only to cite on the next page an engineering conference on structural failure that refused to publish any results. It's an interesting premise, but he goes nowhere (or rather in circles) with it. For a more in-depth and concrete (no pun intended) analysis go to Mario Salvadori, Why Buildings Fall Down.
Anonymous More than 1 year ago
Anonymous More than 1 year ago
Anonymous More than 1 year ago
Anonymous More than 1 year ago
Guest More than 1 year ago
I thought this book was well written and a pleasure to read. I felt it would be a great book to explain to non-engineers in laymen's terms what engineers have to deal with. Engineers in all fields must push the technological envelope and manage risk. This book describes the struggles and why they exist.
Guest More than 1 year ago
This book is a great book on engineering philosophy and the true role of failure in engineering. As in real life, this book points out that we don't learn much from successes, but gain a lot of useful information in our failures that prevent the catastophes from happening again. This is not a mathematical engineering reference book. It is a book that states that when you innovate, mistakes will happen and we must learn from them. Though a little dated, its concepts apply to Challenger, Columbia, and even 9-11 (from the viewpoint of 'why did these happen and how can we keep them from happening again.') I being a recent engineering graduate found it very interesting and read it cover to cover.