The Monster Study on Stuttering, by Wendell Johnson

In 1939, somewhere in the quiet hallways of an orphanage in Davenport, Iowa, a group of children were being told a lie — a carefully constructed, deliberate lie — by a researcher who genuinely believed she was doing science. Some were told that their speech was excellent, fluent, something to be proud of. Others were told the opposite: that they stuttered badly, that people had noticed, that they should stop and think before they opened their mouths. None of them had consented. None of them fully understood what was happening. And several of them would carry the psychological weight of that experiment for the rest of their lives.

The Monster Study, as it came to be known, is one of the most ethically disturbing and scientifically complex experiments in the history of psychology — a study that raises, with brutal clarity, the question of how far researchers are willing to go in the pursuit of a theory they desperately want to prove. It was conceived by Wendell Johnson, a speech pathologist at the University of Iowa, a man who himself stuttered and who had devoted his career to understanding why. Johnson’s central belief — his intellectual obsession, really — was that stuttering is not a biological condition but a learned behavior, one triggered not by anything happening inside the child’s body, but by the reactions of the adults around them. Testing that theory, it turned out, required doing something that his own colleagues would quietly call monstrous.

This article explores the full story: who Wendell Johnson was, what he believed, how the experiment was designed and carried out, what it found, and why it continues to haunt discussions about research ethics, stuttering science, and the vulnerability of children in institutional care.

Wendell Johnson: The Man Behind the Study

To understand the Monster Study, you first need to understand the man at its center — because Johnson’s personal history is inseparable from his scientific agenda. Wendell Johnson was born in 1906 in Roxbury, Kansas, and began stuttering as a young child. Growing up in a time when stuttering was poorly understood and frequently stigmatized, he experienced firsthand the social shame and self-consciousness that the disorder often carries with it. Rather than stepping away from the subject, he moved toward it — studying speech pathology at the University of Iowa, where he would spend nearly his entire career.

Johnson was, by most accounts, an intellectually electrifying figure. He became deeply influenced by General Semantics, a philosophical framework developed by Alfred Korzybski that emphasized the power of language to shape perception and experience. The core idea — that the words we use to describe reality can themselves distort and warp how we experience it — had a profound effect on Johnson’s thinking about stuttering. If language shapes experience, could the simple act of labeling a child a “stutterer” be enough to create the very condition the label described? That question would drive the rest of his career, and ultimately lead to the experiment that would haunt his legacy.

Johnson was also known for his humanity. He cared deeply about people who stuttered, ran a widely respected clinic at the University of Iowa, and published work that was, in many ways, ahead of its time in emphasizing the psychological and emotional dimensions of speech disorders. He was not a villain. He was a man who believed so profoundly in a theory that he was willing to cross ethical lines he perhaps never fully reckoned with — and that tension makes his story genuinely tragic rather than simply scandalous.

The Diagnosogenic Theory: Stuttering as a Learned Label

At the heart of the Monster Study lies Johnson’s most famous and controversial contribution to speech science: the diagnosogenic theory of stuttering. The word “diagnosogenic” essentially means “born from diagnosis” — and that is precisely what the theory proposes. Johnson argued that stuttering does not begin as a neurological or physiological abnormality. It begins, he wrote, “not in the child’s mouth but in the parent’s ear.”

His argument ran something like this: all young children go through a phase of normal speech disfluency. They repeat sounds, stumble over words, pause and restart — this is developmentally typical and, in isolation, nothing to worry about. But some parents, particularly anxious or attentive ones, hear these disfluencies and react with alarm. They label the child a “stutterer.” They correct, worry aloud, ask the child to slow down, start again, breathe. The child — sensing the anxiety, absorbing the label — begins to anticipate difficulty speaking. And it is that anticipation, that self-consciousness, that fear, which produces what everyone then recognizes as actual stuttering. The diagnosis, in other words, creates the condition it claims to describe.

This was a radical, genuinely provocative theory. It had enormous clinical implications — if Johnson was right, the treatment for stuttering was not speech correction but changed attitudes, reduced anxiety, and the removal of stigmatizing labels. It was a theory rooted in compassion and built on real observations. But it was also a theory that needed proof. And Johnson, in 1939, designed an experiment to get it.

The Experimental Design: 22 Children, Two Stories

The experiment was conducted at the Iowa Soldiers’ Orphans’ Home in Davenport, Iowa, in 1939. Johnson supervised the study conceptually and theoretically; the actual hands-on work was carried out by his graduate student, Mary Tudor, who visited the orphanage regularly over a period of several months — from January to late May of that year. The University of Iowa’s established reputation in speech and language research gave the project institutional legitimacy, even as the ethical foundation beneath it was deeply unstable.

Twenty-two orphaned children participated. Their teachers and caregivers had previously identified ten of them as children who stuttered. The remaining twelve were considered normal, fluent speakers. The children were then divided into four groups:

The critical group — the one Johnson’s theory predicted would show the most dramatic results — was Group IIB: fluent-speaking children who were now being systematically told, by an adult authority figure, that they had a speech problem. Tudor visited regularly, delivered the scripted feedback with consistency, and documented what she observed. The orphanage staff, meanwhile, had been briefed to reinforce the same messages. The children had no way of knowing that what they were being told was untrue.

What Happened Inside the Orphanage: Tudor’s Visits

Mary Tudor’s own records from the experiment paint a picture that is genuinely difficult to read. During her sessions, she told the children in the negative feedback groups that she had been listening carefully to their speech and had noticed something concerning. She told the fluent children in Group IIB things like: you have many of the same symptoms as people who stutter — you should try to stop yourself immediately, and think about how you’re talking before you say something. In practice, this meant telling children who had no speech difficulties that they had a problem — and coaching them to become hyperaware, anxious, and self-monitoring every time they opened their mouths.

Even Tudor herself was uncomfortable with what she was doing. Decades later, in interviews given when the study became public, she described feeling deeply uneasy about the sessions — particularly as she began to see behavioral changes in the children she was supposed to help. She reportedly made return visits to some children in the negative-feedback groups to try to undo some of the damage, though these were unofficial and not part of the study protocol. The fact that the person administering the experiment recognized its harm in real time — and still continued — speaks to the profound ethical vacuum in which the study was conducted.

One detail stands out in the historical record as especially troubling: the children in the orphanage were not just experimentally vulnerable — they were structurally vulnerable. As orphans in an institutional setting, they had no parents to advocate for them, no guardians who could withdraw consent, no adult in their daily lives whose primary concern was their wellbeing. They were, in every meaningful sense, a captive population — a fact that made them available for research and simultaneously made any research conducted on them without informed consent a profound ethical violation.

The Results: What the Data Actually Showed

The findings of the Monster Study were, in scientific terms, decidedly mixed — and that complexity is part of what makes the story so rich and frustrating in equal measure. Johnson’s core hypothesis was not cleanly confirmed. The children in Group IIB — the fluent speakers subjected to negative feedback — did not develop full-blown stuttering in the clinical sense of the word. Moments of true stuttering, as strictly defined, did not increase dramatically across the group as a whole.

However. What did increase — measurably and significantly — were speech hesitations, interruptions, self-corrections, and non-stuttering disfluencies. These children became noticeably more cautious, more halting, more self-conscious in their speech. They began to exhibit many of the behavioral hallmarks of someone anticipating and fearing verbal difficulty — even though, at baseline, they had none. Some became quieter. Some began to avoid situations that required them to speak aloud in front of others.

Meanwhile, the children in Group IA — stutterers who received positive, encouraging feedback — showed genuine improvements. Some became more willing to speak in class, more confident, and less avoidant of social situations that had previously triggered anxiety. This was, in its way, a validation of the therapeutic logic underlying Johnson’s theory, even if the full experimental picture was messier than he had hoped. The paradox of the Monster Study is that it managed to cause harm in the pursuit of a theory that, at its humane core, was trying to reduce it.

The Cover-Up: A Study Hidden for Decades

Here is where the story takes another dark turn. Johnson’s colleagues, upon learning of the experiment, were horrified. They nicknamed it the Monster Study — a label that was whispered rather than published, a term of private condemnation rather than public reckoning. And then the study was buried. Johnson, aware of the reputational damage that exposure could cause, never published the research. Tudor’s thesis sat in a university archive. For more than sixty years, the full details of what happened in Davenport in 1939 were known only to a small circle of people in the speech pathology community.

It was not until 2001 that the story became public — when journalist Jim Dyer, then writing for the San Jose Mercury News, uncovered Tudor’s original thesis and published an investigation that drew national attention. The reaction was swift and significant. The University of Iowa issued a formal public apology to the surviving subjects and their families. Several former subjects came forward with accounts of the lasting harm the study had caused them. One woman described having believed her entire adult life that she was a stutterer — a belief instilled entirely by Tudor’s scripted feedback during her childhood — and the revelation that this had been manufactured by an experiment she never consented to was, she said, shattering.

In 2007, six surviving participants and the estates of two deceased subjects filed a lawsuit against the state of Iowa. The case was settled for $925,000 — a monetary acknowledgment that real and lasting damage had been done to real and vulnerable human beings in the name of scientific inquiry.

Long-Term Psychological Impact on the Participants

The legal settlement was important, but it could not fully capture the human reality of what those children experienced. The psychological consequences of the Monster Study were not abstract — they were lived, embodied, and in several cases, lifelong.

Children in the negative feedback groups — particularly those in Group IIB who had been fluent before the study began — showed changes in behavior that persisted well beyond the experiment itself. Some became withdrawn and socially avoidant. Some developed lasting self-consciousness about their speech that shaped their educational experiences, their friendships, their willingness to speak in professional and social settings as adults. The children of at least one original subject reported that their parent had been notably quiet throughout their life and deeply self-conscious about the way they spoke — without ever knowing why.

This is what makes the Monster Study so important as a case study in psychological harm: it illustrates, with painful clarity, how early verbal messaging about identity can shape a person’s self-concept in ways that outlast any formal intervention. The children in the negative feedback groups were not just told they stuttered. They were told, by a trusted adult in a context that felt authoritative, something fundamental about who they were — and for some of them, that message became part of how they understood themselves for decades.

The Ethical Reckoning: What the Study Changed

The Monster Study is now a staple of discussions about research ethics in psychology and medicine — a cautionary example taught in university courses and cited in ethics review guidelines around the world. Its legacy helped reinforce and, in some contexts, accelerate the development of research protections that we now take largely for granted: informed consent, institutional review boards (IRBs), the protection of vulnerable populations, and the principle that the potential harm to participants must always be weighed carefully against the potential scientific benefit of any study.

It also raises questions that don’t have easy answers. Johnson was not operating in a regulatory vacuum entirely of his own making — the ethical norms around research in 1939 were genuinely different from those we hold today. The Nuremberg Code, which established principles for ethical human experimentation, was not formulated until 1947. The Declaration of Helsinki came in 1964. The Belmont Report, foundational to modern American research ethics, was published in 1979. In some sense, Johnson’s study existed in a pre-ethical-framework era — but that explanation, while contextually honest, does not erase the reality that his own colleagues recognized what he was doing as wrong, in the moment, and called it monstrous.

The deeper ethical problem is one of power and vulnerability. Orphaned children in institutional care had no meaningful ability to decline participation, to question what they were being told, or to seek help if they were harmed. The absence of a formal ethics framework didn’t create that vulnerability — it simply failed to protect against it. And that is the enduring lesson the Monster Study carries: that the power differential between researcher and participant must always be treated as an ethical responsibility, not merely a logistical fact.

The Scientific Legacy: Stuttering Research After the Monster Study

For all its ethical darkness, the Monster Study left a complicated scientific legacy — one that the speech pathology community has spent decades trying to honestly assess. Johnson’s diagnosogenic theory, while never cleanly proven by the 1939 experiment, did fundamentally shift the field’s understanding of stuttering. Before Johnson, treatment focused almost entirely on the mechanics of speech: breathing exercises, physical interventions, attempts to retrain the mouth and tongue. Johnson insisted that the mind, the emotions, and the social environment were equally important — and on that, the evidence has consistently supported him.

Contemporary understanding of stuttering is genuinely more nuanced than either side of the 1939 debate could have imagined. We now know that stuttering has neurological and genetic components — brain imaging studies show measurable differences in the speech production networks of people who stutter, and twin studies confirm a significant hereditary element. At the same time, environmental factors, emotional responses, and learned behavioral patterns clearly influence the severity and expression of the disorder. The truth, as it turns out, is that stuttering begins both in the child’s mouth and in the parent’s ear — and in the brain, and in the culture, and in the complex interplay between all of them.

Patricia Zebrowski, a University of Iowa speech pathology professor, noted that despite its profound ethical failures, Johnson’s research produced what remains “the largest collection of scientific information” on stuttering at the time, and that his insistence on the importance of the stutterer’s own thoughts, feelings, and attitudes toward their speech continues to influence treatment approaches today. Cognitive-behavioral therapy for stuttering, mindfulness-based approaches, and the emphasis on acceptance and identity in modern fluency programs all carry Johnson’s intellectual fingerprints — even as his methods remain indefensible.

What the Monster Study Teaches Us About Psychological Vulnerability

Step back from the historical specifics for a moment and the Monster Study becomes something even larger than the story of one flawed researcher and twenty-two unlucky children. It becomes a window into something deeply important about how psychological vulnerability works — and why protecting it matters so profoundly.

The children in Group IIB did not stutter. And yet, when an authority figure told them they did — told them repeatedly, in a context designed to feel serious and clinical — many of them began to behave as though they did. They became self-conscious, avoidant, hesitant. They internalized the label and made it part of how they moved through the world. This is not weakness. This is how human beings work. We are, all of us, profoundly shaped by what we are told about ourselves — especially in childhood, especially by people we perceive as authorities, and especially when we have no other adult in our lives to offer a counternarrative.

This is why the way we speak to children about their minds, their behaviors, and their perceived difficulties matters so enormously. It is why language around mental health, learning differences, and behavioral challenges must be chosen with care, compassion, and precision. Labeling is not a neutral act. It carries weight. And the Monster Study demonstrates, with a specificity that no theoretical argument could match, exactly how much.

If you or someone you love is navigating a speech disorder, communication anxiety, or any challenge that has become intertwined with self-identity and self-worth, it is worth knowing that those experiences are deeply human — and that specialized, compassionate support exists. Reaching out for that support is not a sign of weakness. It is, in every sense, an act of strength and self-advocacy.

FAQs About The Monster Study on Stuttering

Why was it called the Monster Study?

The name was not a formal title Johnson chose — it was a term coined by Johnson’s own colleagues at the University of Iowa when they learned what he had done. They were horrified by his willingness to conduct experiments that might cause psychological harm to orphaned children, and “the Monster Study” became the informal, whispered name for the research within academic circles. The name stuck, and when journalist Jim Dyer brought the study to public attention in 2001, it became widely used to describe the experiment in both academic and popular discourse. Johnson himself never used the term publicly and kept the study unpublished throughout his lifetime, apparently aware of the reputational damage that full disclosure would cause.

Did the Monster Study actually cause children to stutter?

This is one of the most technically contested questions surrounding the study. In strict clinical terms, the negative-feedback group of fluent children did not develop what speech pathologists would classify as persistent stuttering. However, they did exhibit a measurable increase in speech hesitations, interruptions, self-corrections, and avoidance behaviors — all of which are associated with stuttering-related anxiety and self-consciousness. Some children became noticeably quieter and more withdrawn in situations requiring speech. Whether these changes constitute “induced stuttering” depends partly on how stuttering is defined, but the psychological and behavioral harm was real and documented, regardless of what label is applied to it.

Was Wendell Johnson ever held accountable for the Monster Study?

Johnson died in 1965, long before the study became public knowledge, so he never faced formal professional or legal accountability. The University of Iowa, however, issued a public apology in 2001 when investigative journalist Jim Dyer exposed the research. In 2007, six surviving participants and the estates of two deceased subjects filed a lawsuit against the state of Iowa, which was settled for $925,000. The university acknowledged that the experiment had been conducted unethically and expressed regret for the harm caused to the participants. Johnson’s academic reputation remains a complex one — recognized for genuine contributions to the understanding of stuttering, but permanently shadowed by the ethical failures of the 1939 study.

What is the diagnosogenic theory and is it still relevant today?

The diagnosogenic theory, proposed by Johnson in the early 1940s, holds that stuttering is primarily a learned behavior triggered not by any neurological abnormality but by the negative reactions of adults — especially parents — to normal childhood speech disfluencies. When a child is labeled a “stutterer” and responds with anxiety and self-monitoring, that anxiety itself produces the disfluent speech that then confirms the label. While modern research has established that stuttering has significant neurological and genetic components that Johnson underestimated, his core insight — that psychological factors, social feedback, and self-perception profoundly shape the experience and severity of stuttering — remains influential and clinically relevant in contemporary speech therapy approaches.

What ethical safeguards exist today to prevent studies like this?

The development of formal research ethics frameworks in the decades following World War II — including the Nuremberg Code (1947), the Declaration of Helsinki (1964), and the Belmont Report (1979) — established the core principles that now govern human research: informed consent, beneficence, non-maleficence, and justice. In the United States, all research involving human subjects at accredited institutions must be reviewed and approved by an Institutional Review Board (IRB), which evaluates potential risks against anticipated benefits and ensures that participants are fully informed before agreeing to take part. Particular protections apply to vulnerable populations including children, prisoners, and people with cognitive or developmental conditions. Research like the Monster Study would not receive approval under any contemporary ethical review process.

How did Mary Tudor, the researcher who administered the study, reflect on it later in life?

Mary Tudor gave rare on-the-record interviews when the study became public in 2001, at which point she was in her eighties. By multiple accounts, she had carried significant regret about her role in the experiment for decades. She reportedly made unofficial return visits to some of the children in the negative-feedback groups during the original study period, attempting informally to undo some of the harm she had caused — though these efforts were not part of the formal protocol and are difficult to assess in terms of their effectiveness. Her willingness to speak about her experience, even so many years later, contributed meaningfully to the historical record and to the broader understanding of how researchers can find themselves complicit in harm even when acting in good faith within a flawed system.

Is stuttering now well understood as a condition?

Significantly better understood than in Johnson’s era — though still not completely. Modern neuroscience has identified measurable differences in the brain structure and function of people who stutter, particularly in the networks governing speech planning and motor execution. Genetic research has identified several gene variants associated with elevated stuttering risk. At the same time, research confirms that psychological factors — anxiety, avoidance, self-perception, and the social environment — play a powerful role in how stuttering is experienced and expressed. Treatment today is multidisciplinary, often combining speech-language therapy techniques with cognitive-behavioral approaches that address the emotional and identity dimensions of the disorder. The field has moved well beyond the binary of “biological” versus “learned” — toward a more honest acknowledgment of how deeply intertwined those forces always are.