The We Hunted the Mammoth Pledge Drive continues! If you haven’t already, please consider sending some bucks my way. (And don’t worry that the PayPal page says Man Boobz.) Thanks! And thanks again to all who’ve already donated.

The ManKind Initiative, a UK organization devoted to fighting domestic violence against men, recently put out a video that’s been getting a lot of attention in the media and online, racking up more than six million views on YouTube in a little over a week.

The brief video, titled #ViolenceIsViolence, purports to depict the radically different reactions of bystanders to staged incidents of domestic violence between a couple in a London plaza. When the man was the aggressor, shoving the woman and grabbing her face, bystanders intervened and threatened to call the police. When the woman was the aggressor, the video shows bystanders laughing, and no one does a thing.

The video has been praised by assorted Men’s Rights Activists, naturally enough, but it has also gotten uncritical attention in some prominent media outlets as well, from Marie Claire to the Huffington Post.

There’s just one problem: The video may be a fraud, using deceptive editing to distort incidents that may well have played out quite differently in real life.

A shot-by-shot analysis of the video from beginning to end reveals that the first “incident” depicted is actually a composite of footage shot of at least two separate incidents, filmed on at least three different times of day and edited together into one narrative.

A careful viewing of the video also reveals that many of the supposed “reaction shots” in the video are not “reaction shots” at all, but shots taken in the same plaza at different times and edited in as if they are happening at the same time as the staged “incidents” depicted.

Moreover, none of the people depicted as laughing at the second incident are shown in the same frame as the fighting couple. There is no evidence that any of them were actually laughing at the woman attacking the man.

The editing tricks used in the video were brought to my attention by a reader who sent me a link to a blog entry by Miguel Lorente Acosta, a Professor of Legal Medicine at the University of Granada in Spain, and a Government Delegate for Gender Violence in Spain’s Ministry of Equality. He goes through the video shot by shot, showing each trick for what it is.

The post in Spanish, and his argument is a little hard to follow through the filter of Google Translate, so I will offer my own analysis of the video below, drawing heavily on his post. (His post is still worth reading, as he covers several examples of deceptive editing I’ve left out.)

I urge you to watch the video above through once, then follow me through the following analysis.

The first “incident” is made up of footage taken at three distinct times, if not more. The proof is in the bench.

In the opening shot of the video, we see an overview of the plaza. We see two people sitting on a bench, a man in black to the left and a woman in white to the right, with a trash can to the right of them. (All of these lefts and rights are relative to us, the viewers.) The trash can has an empty green bag hanging off of it.

As the first incident begins, we see the same bench, only now we see two women sitting where the man was previously sitting. The trash can now has a full bag of trash sitting next to it.

In this shot, showing bystanders intervening in what is portrayed as the same fight, and supposedly depicting a moment in time only about 30 seconds after the previous shot, we see that the two women on the bench have been replaced by two men, one in a suit and the other in a red hoodie. The full trash bag has been removed, and the trash can again has an empty trash bag hanging off of it.

Clearly this portion of the video does not depict a single incident.

What about the reaction shots? The easiest way to tell that the reaction shots in the video did not chronologically follow the shots that they come after in the video is by looking at the shadows. Some of the video was shot when the sky was cloudy and shadows were indistinct. Other shots were taken in direct sunlight. In the video, shots in cloudy weather are followed immediately by shots in roughly the same location where we see bright sunlight and clear shadows.

Here’s one shot, 9 seconds in. Notice the lack of clear shadows; the shadow of the sitting woman is little more than a vague smudge.

Here’s another shot from less than a second later in the same video – the timestamp is still at 9 seconds in. Now the plaza is in direct sunlight and the shadows are sharp and distinct.

If you watch the video carefully, you can see these sorts of discontinuities throughout. It seems highly unlikely that the various reaction shots actually depict reactions to what they appear to be reactions to. Which wouldn’t matter if this were a feature film; that’s standard practice. But this purports to be a depiction of real incidents caught on hidden camera and presented as they happened in real time.

The issue of non-reaction reaction shots is especially important when it comes to the second incident. In the first incident, we see a number of women, and one man, intervening to stop the violence. There is no question that’s what’s going on, because we see them in the same frame as the couple.

In the second incident, none of the supposed laughing onlookers ever appear in the same frame as the fighting couple. We have no proof that their laughter is in fact a reaction to the woman attacking the man. And given the dishonest way that the video is edited overall, I have little faith that they are real reaction shots.

The people who are in frame with the fighting couple are either trying resolutely to ignore the incident – as many of the onlookers also did in the first incident – or are clearly troubled by it.

I noticed one blonde woman who looked at first glance like she might have been laughing, but after pausing the video it became clear that she was actually alarmed and trying to move out of the way.

There is one other thing to note about the two incidents. In the first case, the onlookers didn’t intervene until after the man escalated his aggression by grabbing the woman by her face. In the second video, the screen fades to black shortly after the woman escalates her aggression to a similar level. We don’t know what, if anything, happened after that.

Is it possible that the first part of the video, despite being a composite of several incidents, depicts more or less accurately what happened each time the video makers tried this experiment? Yes. Is it possible that onlookers did indeed laugh as the woman attacked the man? Yes.

But there is only one way for The ManKind Initiative to come clean and clear up any suspicion: they need to post the unedited, time-stamped footage of each of the incidents they filmed from each of their three cameras so we can see how each incident really played out in real time and which, if any, of the alleged reactions were actual reactions.

In addition to the editing tricks mentioned above, we don’t know if the video makers edited out portions of the staged attacks that might have influenced how the bystanders reacted.

The video makers should also post the footage of the incidents that they did not use for the advert, so we can see if reactions to the violence were consistently different when the genders of attackers and victims were switched. Two incidents make up a rather small sample – even if one of these incidents is actually two incidents disguised as one.

Domestic violence against men is a real and serious problem. But you can’t fight it effectively with smoke and mirrors.