Skip to main content
Category: Laws
Type: Popular maxim about missing instructions
Origin: Chinese management compilations. No dated text by a verified B. Gilbert was found.
Also known as: 吉尔伯特法则 (Guilbert Rule). Not the separate entry that says people like to work for people they like.
Quick Answer — Gilbert’s Law is the circulating claim that the surest signal of a work crisis is that nobody tells you how the work should be done. The encyclopedia gloss reads that silence as neglect: if nobody corrects you, you have been left outside the rules. A second, English, circulation says that once the job is yours, the method is yours. This page keeps both, and uses a narrower split. The crisis is that nobody has stated the result and the boundary. A mediocre system fills the silence with mechanical motion orders. A better system states what “done” is, then requires the person to define the problem and create the method. Silence about errors is still a failure. It is not the same failure as the absence of a motion script.

​
What is Gilbert’s Law?

Gilbert’s Law is a popular maxim about a gap at work. The gap is that nobody has said how the work should proceed. What you do with that gap is the whole argument.
The surest signal of a work crisis is that nobody tells you how it should be done.
That sentence is this page’s translation of the line printed by the MBA智库百科 entry read here: 工作危机最确凿的信号，是没有人跟你说该怎样做。 The entry names the proposer as 英国人力培训专家 B·吉尔伯特 and spells the English label Guilbert Rule. It adds a second line, which this page translates as: the truly dangerous thing is that nobody talks with you about the danger. The entry does not print a year, a speech, or a book page in the portion read here. The name links to an empty wiki page. This page does not treat either line as a recovered quotation. The same silence is read in two incompatible ways. The compilation’s own gloss says that correction is care. When a boss or a parent stops saying what you did wrong, you have not become free. You have been dropped. English workplace pages circulate a different sentence, “The biggest problem at work is that no one tells you what to do,” and often read it as initiative: finding the method is your job. “How” in the Chinese line and “what” in the English line are not the same word. A 2026 explainer, read in excerpt, says no authoritative source names an inventor or a year. Some posts name Gilbert Lafayette Laws, or the psychologist Thomas F. Gilbert. Neither name is on the MBA entry. No primary document for those attributions was opened. Do not cite either man as the author of the sentence. The usable split is this page’s reading, not a third quotation. Ask whether anyone has stated the accomplishment and the boundary: what “done” is, and what must not happen. If the answer is no, that is the crisis in the maxim. Filling the gap by dictating every motion manages the behavior and still has not named the result. Naming the result, then requiring the person to define the problem and invent the method, is the other response. Productive Thinking Model is a separate method that also starts by stating the problem. It is not a second name for this sentence. Thomas F. Gilbert (1927–1995) is the researched neighbor, and only a neighbor. The Wikipedia summary read for this page calls him a psychologist, often described as a founder of human performance technology, who used the term performance engineering. His book is Human Competence: Engineering Worthy Performance. The summary says formal learning often changes knowledge without changing behavior, that he treated performance as depending on both behavior and the environment, and that he believed the greater barrier to exemplary performance was missing support in the work setting, not a lack of knowledge or skill. He is said to have looked at the environment before the person. The 1978 book was not re-opened. The summary does not contain the maxim. An often-repeated percentage for how many gaps sit in the environment is not printed here.

​
Gilbert’s Law in 3 Depths

  • Beginner: If nobody has said what “done” looks like, you do not yet have the job. A list of motions is not that sentence.
  • Practitioner: Write the result, the boundary, and the error someone will still name. Then design the method. Do not wait for a script, and do not treat silence about errors as freedom.
  • Advanced: “Nobody corrects me” and “nobody handed me the steps” are different gaps. A sticky name is not a source. A course that only adds knowledge is not the first repair the neighbor literature points to, and it is not proof of this maxim.

​
Origin

Gilbert’s Law circulates as a named sentence in Chinese management encyclopedias. The MBA智库百科 entry read for this page prints the crisis line, the gloss that correction is care, the label Guilbert Rule, and the office “British manpower-training expert.” It prints neither a year nor a place where the sentence was said. Stigler’s Law of Eponymy is the neighboring warning. A name that sticks is a handle. It is not, by itself, a record of who said the sentence first. A different MBA entry uses the Chinese title 吉尔伯特定律 for another sentence: people like to do things for people they like. That entry, read in excerpt, names an American manager, 瑟夫·吉尔伯特. It was not read line by line for this article. The two sentences are not aliases. Stories about liking, praise, or ancient generals belong to that other entry. They are not evidence for the crisis line. The English sentence about nobody telling you what to do travels without the Chinese gloss. Initiative pages treat the gap as your duty. Neglect pages treat the gap as abandonment. Both can cite “Gilbert” and describe opposite duties. The split on this page is how those two readings can both be partly right. Abandonment is silence about the result and about the danger. Duty begins after the result and the boundary have been stated, when the method is still unwritten. The performance-engineering neighbor is not a signature on either sentence. Wikipedia’s summary of Thomas F. Gilbert, read here, places his book in a publication line dated 1978 and 2007, Pfeiffer, San Francisco, tied to the International Society for Performance Improvement. The same summary says the society’s professional award was renamed for him in 1996. Ogden Lindsley’s memoriam in The Behavior Analyst (1996) is cited there and was not opened. Biographical lines on that summary, including degrees and a year with B. F. Skinner, are not repeated here, because the Wikipedia article itself is flagged as short of citations. What this page uses is the distinction the summary does state: support in the setting versus knowledge in the person, and behavior versus the performance that setting is supposed to produce.

​
Key Points

Four distinctions keep the maxim from turning silence into either a scolding or a blank check.
1

The signal is an unstated result, not a mood

The compilation’s sentence is about a missing instruction: nobody tells you how the work should be done. It is not a claim that worry, or a quiet office, is itself the crisis. Until someone can say what “done” is, and what must not happen, the job has not been stated. A feeling of freedom does not replace that sentence.
2

Neglect and an unwritten method are different gaps

The compilation reads silence as being dropped: nobody points out the error, so the danger has no name. The initiative reading says the method was never going to arrive in a memo, so you must make it. Ask which gap you are in. A person can have a clear result and still have no script. A person can have a thick script and still not know what the result is, or which error anyone will name.
3

A motion script is the mediocre filling

Dictating every motion answers the silence by managing behavior. The result can stay unstated. The better filling states the accomplishment and the boundary, then requires the person to define the problem and create the method. Galls’ Law is a neighboring warning about specifying a complex working system from nothing. It is not this maxim. Creating the method does not mean inventing a new aim. The aim was supposed to be the sentence someone did say.
4

Silence about the danger is still a failure

The compilation’s second line keeps correction in the system. Autonomy about sequence is not permission to hide an error, and it is not a reason for a lead to stop naming danger. If nobody will say what is going wrong, the Chinese gloss applies: that silence is the dangerous one. A review of the result is part of the better system. A review that only checks whether your motions matched an old script is the mediocre system again.

​
Applications

Use the maxim to separate an unstated result from an unwritten method. Do not use it as a slogan for either more rules or no rules.

Write done before you start

Before the work, write one sentence for the result and one for the boundary. If you cannot, ask for those two sentences. The question is not a request for every motion. If the answer is still silence, the crisis in the maxim is still open.

Hand over the result, not the choreography

When you lead, state the accomplishment and the constraint. Ask the person to define the problem and propose the method. Review the result. Resemblance to your old motions is not the result. Constraint-Based Thinking is a way to name the limit. It is not a script.

Keep a name for the error

Agree, in advance, how a miss will be said. The compilation treats a workplace where nobody mentions danger as the dangerous one. Praise is a different encyclopedia entry. Correction of a named miss is this one.

Sort the silence in the meeting

When someone says they were not told what to do, ask which sentence was missing: the result, the boundary, or the motion list. Only the first two are this page’s crisis. The third may be the method they still have to create.

​
Case Study

The counted neighbor is not a test of the maxim. It is a summary of job features and ratings, and it shows why a single percentage cannot be carried onto a team that “stopped giving instructions.” Stephen E. Humphrey, Jennifer D. Nahrgang, and Frederick P. Morgeson, “Integrating motivational, social, and contextual work design features: A meta-analytic summary and theoretical extension of the work design literature,” Journal of Applied Psychology 92, number 5, 2007, pages 1332–1356. DOI 10.1037/0021-9010.92.5.1332. PMID 17845089. The PDF was not opened. The abstract below is the one printed on the University of Iowa repository page for that article. That page lists Humphrey at Florida State University, and Nahrgang and Morgeson at Michigan State University.
Results from a summary of 259 studies and 219,625 participants showed that 14 work characteristics explained, on average, 43% of the variance in the 19 worker attitudes and behaviors examined.
The same abstract’s examples: motivational characteristics explained 25% of the variance in subjective performance, 2% in turnover perceptions, 34% in job satisfaction, 24% in organizational commitment, and 26% in role perception outcomes. Beyond those, social characteristics explained incremental variances of 9% in subjective performance, 24% in turnover intentions, 17% in job satisfaction, 40% in organizational commitment, and 18% in role perception outcomes. Beyond both, work-context characteristics explained incremental variances of 4% in job satisfaction and 16% in stress. “Explained” is the abstract’s word for a share of variance in that summary. It is not an experiment that assigned one workplace to this maxim. Turnover perceptions (2%) and turnover intentions (24%, and incremental) are different lines. They are not one turnover effect. The abstract does not split the 25% into autonomy, feedback, or any other single feature. This page therefore does not say that freedom of method explained 25%, or that naming a result explained 25%. The 43% is an average across 19 outcomes, not a performance score. Do not export 259, 219,625, 43%, 25%, 34%, 24%, 26%, 2%, 9%, 17%, 40%, 18%, 4%, or 16% onto a shop, a class, a household, or another year. Do not read them as the gain from deleting a checklist. A worked pass, not a study, shows the split the percentages cannot. A lead says, “Stand here, then here, then seal the crate,” and never says what the crate must satisfy. The silence about the result remains. Another lead says: the order is done when every line is in the crate and the exception list is empty; you choose the sequence; if a line will not fit, you say so before the truck leaves. The second version states the accomplishment and the boundary, and leaves the method. Nobody in that paragraph was observed. It is a diagram of the sentence, not a result from the 2007 summary. The MBA entry also attaches a 1986 shuttle retelling, including a printed gasket thickness of 28 inches, about 85 cm. That thickness is not used as history. The flight figures on that entry are not repeated here. A disaster narrative is not a measure of this maxim.

​
Boundaries and Failure Modes

Gilbert’s Law fails when the method is asked to supply a result nobody stated, or when a variance share is told as the effect of deleting instructions. The first boundary is the kind of error. Where a wrong motion is fatal or irreversible, the boundary has to be specific, and “create the method” does not mean “invent the safety step.” The compilation’s second line still applies: somebody has to be willing to name the danger. Medicine, flight, and a live money movement are in this class. The page does not set the class with a statute. It only refuses the reading that every unscripted act is the mature form of the maxim. The second boundary is the 2007 abstract. Fourteen characteristics, taken together, track ratings in that summary. The abstract does not identify which characteristic is “a motion script” or “a stated result.” A team that removes instructions and keeps the old ambiguity has not performed the split on this page. Peter Principle is about a promotion past the level of competence. Dilbert Principle is a different named claim, about where incompetent work is parked. Neither is this sentence. The third boundary is Thomas Gilbert’s neighbor status. Wikipedia’s summary says he believed missing performance support was the greater barrier, and that the environment should be examined before the person. It does not report a percentage, a trial, or this maxim. A knowledge course can still be the right repair when the person cannot yet do the method. The summary’s own starting point is that a course which changes only knowledge may leave behavior unchanged. That is a warning about training as the first lever. It is not a finding that instruction is useless. The common misuse is to hear one silence and prescribe the other cure. “Nobody told me the steps” is answered, on this page, by stating the result and writing the method. “Nobody told me the steps” is not answered by a thicker motion list that still hides the result. “Nobody corrects me” is not answered by calling the quiet a mark of trust.

​
Common Misconceptions

The name travels with additions that the sources read for this page do not support.
The encyclopedia names B·吉尔伯特 and an English spelling, Guilbert Rule. The name links to an empty page. No year, speech, or book page was in the portion read here. Use the sentence as a circulating maxim. Do not cite a trainer, a city, or a publication date.
Thomas F. Gilbert wrote Human Competence (1978). He is the performance-engineering neighbor summarized above. He is not the signature on the MBA line. Posts that name Gilbert Lafayette Laws were not matched to a document opened for this page. Do not promote either attribution from a blog byline.
A separate encyclopedia entry titled 吉尔伯特定律 says people like to do things for people they like, and names 瑟夫·吉尔伯特. That sentence is not this page. Praise, liking, and general stories are not the crisis line. Dilbert Principle is also a different page. The surnames only look alike in English.
These pages mark the neighbors. None of them is a second name for Gilbert’s Law.

Dilbert Principle

A claim about where weak work is placed. The shared English look of the names is not a shared sentence.

Peter Principle

Promotion past the level of the work. This page is about whether the result was stated, not about the rank.

Stigler's Law of Eponymy

A sticky name is a weak record of who said a sentence first. That is why the trainer on the encyclopedia line is not used as a source.

Productive Thinking Model

A method that starts by stating the problem before the moves. It is not this maxim, and it is not a motion script.

Constraint-Based Thinking

Name the limit that the result has to respect. A limit is a boundary, not a list of motions.

Galls' Law

A complex system that works was not specified in full on the first day. That warning is adjacent. It does not replace a stated result.

​
One-Line Takeaway

The crisis is an unstated result and an unnamed danger. State both, then require the person to define the problem and create the method. A list of motions fills neither gap.