Zeke, Zach and Zebediah are co-workers and Facebook friends. One day at work, Zach pulls out his smartphone, accesses Facebook and writes an offensive comment about Zebediah’s religion on Zebediah’s wall. Zebediah has received several offensive comments at work about his religion and he has reported it to no avail. Zeke, also at work, logs on to Facebook on his work computer, sees the offensive comment and clicks “like”. Zebediah, working from home, also logs in and sees the offensive comment and “like”. Zach’s comment might be harassment but what about Zeke “liking” it?

Although social media is not a novel opportunity to offend others, the issue of social media-based harassment in the employment context is still relatively unexplored. Rather than identify a single unexplored problem, the approach here selects a marginal piece of social media activity and pulls on that thread to survey the disconcerted legal status quo that applies to it.

The thread tugged here is the Facebook “like”, a simple way for one user to approve content posted by another, resulting in a statement that the user “likes” the content, which is then added to the “liking” user’s Facebook wall where others can view it. Facebook is not alone in providing these simple assent-based actions. Google Plus adopted the “+1” internet meme as its version of the “like”. Other sites allow you to give content a thumbs up or thumbs down (for example, YouTube). Additionally, many sites allow users to republish content with little effort. For example, Facebook allows users to share content in a variety of channels on the site and Twitter also allows users to retweet with a click. The Facebook “like” will be used as a proxy for all of these options because it is easily understood and well known. While a “like” may seem facially insignificant, it does more than just show approval. It turns a single offensive comment into group activity and spreads the offensive content by making it available to Facebook friends of the “liking” user, increasing the audience.

The analysis will briefly assess each stage of inquiry whether a “like” can form part of a hostile work environment. The analysis begins with a brief overview of hostile work environment claims, followed by a review of the competing federal circuit approaches to what is workplace activity. The discussion then focuses on two decisions addressing social media-based harassment and the unanswered questions. After addressing the work environment part of a hostile work environment claim, the analysis turns to the hostile component and reviews judicial approaches to workplace events analogous to social media. Recognizing no analogous situation completely fits the “like”, the analysis addresses two cases in different labor and employment context addressing “likes” and how those rules could apply. Finally, a synthesis of approaches is suggested as a compromise between the status quo and the need to resolve the unanswered questions along the way.

I.                   Overview of Hostile Work Environment Claims

Hostile work environment claims may be asserted for any protected status under federal and state anti-discrimination laws. Hostile work environment claims require the plaintiff to prove the workplace conduct was: (1) severe or pervasive; (2) unwelcome; (3) rendered plaintiff’s working environment objectively and subjectively hostile sufficiently to alter the plaintiff’s working conditions; and the conduct was because of plaintiff’s protected status.^[1] A plaintiff must also prove a basis for imputing liability on the employer. In a hostile work environment claim where a supervisor’s conduct results in a tangible employment action, such as a failure to promote, the employer is strictly liable for the supervisor’s conduct. If there is no tangible action but a supervisor is responsible for a hostile work environment, liability will be imputed on the employer unless it successfully asserts the Ellerth affirmative defense. Under this affirmative defense the employer must prove it: (1) acted reasonably to prevent and correct harassment; and (2) plaintiff failed to take advantage of corrective opportunities.^[2] In a hostile work environment created by co-workers, liability is imputed under a negligence standard, so a plaintiff must prove the employer had actual or constructive knowledge of the harassment and failed to take effective action.^[3]

To determine whether the conduct was severe or pervasive, courts will look at the totality of the circumstances. Courts consider several factors, including: (1) frequency; (2) severity; (3) whether the harassment was physically threatening, humiliating, or merely an offensive utterance; (4) whether the harassment unreasonably interfered with plaintiff’s work performance; (5) whether the conduct was hostile and patently offensive; (6) whether the harasser was a supervisor or co-worker; (7) whether others joined; and (8) whether it was directed at more than one individual.^[4] A Facebook “like” will implicate several of these factors.

Although several factors may suggest a “like” is evidence of a hostile work environment, the existing case law makes that conclusion uncertain. Limited case law exists on the direct question of social media in a hostile work environment and no court has addressed the finer details of this particular social media activity. Adding further complexity, courts addressing social media in other labor and employment law contexts have taken widely different approaches to the relevance of social media and in particular, the “like”.

II.                Addressing the “Like”

A.    What’s Not to “Like”

The “like” is itself a unique piece of evidence; it lacks independent meaning and must be associated with an underlying comment “liked” to determine its significance. The “like” only indicates asset to another’s comment. It is analogous to one employee making a sexist comment and another employee walking by and giving the commenter a thumbs up or laughing about it. However, unlike a passing thumbs up or chuckle, a “like” is permanent and may be public. In that regard, it may be a more meaningful act.

Additionally, several issues arise regarding how the comment and “like” relate to the victim and the workplace. The comment and “like” may be on an employee’s Facebook wall, similar to a private conversation at work overheard by the victim. One co-worker may put an offensive comment on the victim’s page, analogous to comments made directly to the victim. Alternatively, the “like” may apply to a comment by an unrelated third party but seen by the victim. That is analogous to the victim seeing an email exchange on a co-worker’s computer with an offensive statement or picture. Each of the analogous non-social media situations could be evidence of a hostile work environment but the courts have not expressed a clear rule that liability necessarily arises in any of them. All of these analogies to non-social media interactions rely upon the underlying assumption the Facebook interactions occur within the workplace. Courts similarly do not agree that is the case.

B.     Facebook and the Workplace

Whether social media is a component of the workplace is itself a conundrum. In the modern workplace, many employees can access social media through employer devices or their own. What may normally be outside activity bleeds into the physical workplace through these devices. The Facebook comment, “like” and view by the victim may all or individually occur inside the physical workplace. Courts disagree where the boundaries of workplace activity exist.

A split exists between federal circuits whether activity outside the workplace is evidence of a hostile work environment.^[5] In circuits limiting evidence to activity within the physical workplace, the admissibility of a “like” is less likely than those adopting more liberal views. While workplace boundaries are frequently drawn in physical terms, that definition leaves the intersection of physical workplace and internet undefined. Activity may occur physically within the workplace but on non-employer equipment. Is it workplace activity if the “like” is made at work on a personal smartphone? Is it workplace activity if “liked” at work but only seen by the victim after work at home? Does it matter if the underlying comment is a spillover from workplace activity? When an employee works from home and engages in discriminatory social media activity, is that workplace activity? Inevitable invitations will arise to address these questions.

The Supreme Court of New Jersey provided a framework for analyzing the issue in Blakey v. Continental Airlines. Blakey was a female pilot who filed a charge of sex discrimination and retaliation due to gender-based comments and pornography displayed in the cockpits by male pilots. While litigating her claims, several male pilots posted comments criticizing her on an online message board hosted by a third party on behalf of Continental. The message board was available through Continental’s computers and employees’ own computers at home. Although Continental did not operate the message board or formally monitor it, Continental had access to it and its employees voluntarily monitored and provided technical support.^[6]

The court rejected Continental’s contention that the message board was not part of the workplace. Holding the messages were sufficiently severe had they been on a physical bulletin board in a Continental worksite, the court addressed whether the online message board was part of the workplace. Applying New Jersey precedent, it held activity beyond the physical workplace can be actionable when it permeates the workplace. Where work-related, off-duty harassment continues a pattern of workplace harassment, the employer sends the message harassment is acceptable if it does not take steps to eliminate the off-duty harassment.^[7]  

The test the court provided on remand to determine if the message board was part of the workplace was whether Continental obtained a substantial workplace benefit from it. If so, then the content can be considered workplace activity. Probative factors include the number of employee-users and benefits to the efficiency and operations of the business.^[8] At least one commentator has suggested this test draws an effective distinction for off-duty social media harassment while recognizing the test is limited to conduct within an employer’s own social media portals, such as its own Facebook page or an internal social media system, but would not resolve conduct between employees on their personal social media accounts.^[9]

Another court has addressed harassment between co-workers on personal Facebook accounts. In Amira-Jabbar v. Travel Services, Inc. the Puerto Rico District Court considered a picture comment by a co-worker as evidence of a hostile work environment.^[10] Plaintiff Amira-Jabbar, a black woman, asserted a hostile work environment based on actions by co-workers. One instance Amira-Jabbar relied on related to pictures taken at an employer-approved outing posted by Marilyn Hernandez, a co-worker, to her Facebook account. Amira-Jabbar added a comment to a picture stating, “…remind me that taking pictures in the shade is really a disservice to my wonderful chocolate skin.” A third co-worker, Miguel Hernandez, subsequently commented, “[t]hat is why you always have to smile!!!” Amira-Jabbar claimed the response was racist.^[11]

The court accepted the Facebook comment as part evidence of a hostile work environment but dismissed the claim based on the employer’s affirmative defense. The court did not analyze whether Facebook content was within the workplace, it simply accepted Amira-Jabbar’s inclusion as evidence of workplace harassment. When informed of the comment, Travel Services blocked access to Facebook at work. The court held this was a sufficient response to avoid liability for co-worker harassment.^[12]

The outer boBy moduleundaries of the employer’s duty in Amira-Jabbar appear to be the physical workplace and preventing access to offensive content. The opinion does not indicate if any of the Facebook activity occurred or was viewed at work but the content was still available beyond the physical workplace to the offender, victim and likely co-workers. The court did not address how the continued, off-duty availability of the offending reply might continue to permeate the workplace even though it could not be viewed at work on the employer’s computers.

If Blakely represents the liberal approach to defining the workplace, Amira-Jabbar represents the narrow approach. There is a lot of room in between for courts to stake a position. Perhaps a synthesis of the two standards is the best approach: apply Blakely’s substantial workplace benefit test when the employer controls the forum but Amira-Jabbar’s restricted employer duty when the employer does not. However, Amira-Jabbar fails to create effective guidance when employees harass through their own social media accounts. Puerto Rico is within one circuit that holds outside activity can be part of a hostile work environment^[13]; however, the Amira-Jabbar court seemed unconcerned with the Facebook content once the employer blocked it from the physical workplace. Amira-Jabbar’s co-workers were seemingly free to harass her on Facebook so long as the employer could not see it. A ­Blakely­-­Amira-Jabbar approach fails to address the concern expressed in Blakely that off-duty conduct permeates the workplace. This unaddressed concern is compounded when employees can use Facebook on personal devices on their own accounts during work hours in the workplace.

C.    The Significance of the “Like”

If a court decides a “like” is workplace activity it may not be proof of a hostile work environment. Not all offensive conduct in the workplace is probative to a hostile work environment claim. Isolated incidents are rarely evidence on their own of a hostile work environment, making the “like” and underlying comment unlikely to prove a plaintiff’s case alone. Furthermore, courts addressing harassment claims deemphasize harassing statements when sent through electronic channels and where the conversations are brief, suggesting a “like” has little probative value. However, none of the prior opinions dealing with similar issues have specifically addressed “liking” offensive content and how that social media platform may alter the significance of such a minor act. Blakely and Amira-Jabbar both address some of the complexity of the intersection of social media and the physical workplace but these cases leave unanswered the importance of specific social media conduct. While the prior section addressed whether a “like” could be part of the workplace, this section will address the value courts may assign it if it is.

While an individual statement by a supervisor would be subject to admissibility analysis under the stray comments doctrine^[14], an individual comment in a hostile work environment claim is more likely to be analyzed for its substantive merits because a hostile work environment usually arises from an accumulation of otherwise stray comments or isolated incidents. In assessing the factors discussed in the first section to determine whether conduct is severe or pervasive, isolated incidents are insufficient unless extremely serious, such as incidents involving nonconsensual touching, because isolated incidents by themselves rarely affect the terms and conditions of employment.^[15] Even a handful of sporadic or isolated incidents generally fail to prove a hostile work environment.^[16] Therefore it is improbable a single “like”, even when paired with the underlying content, would be sufficient on its own.

Additionally, courts tend to deemphasize electronic evidence of harassment. For example, the Eastern District of Virginia in Rivera v. Prince William County School Board held an email containing a request for sexual conduct and nonconsensual sexual touching was neither severe nor pervasive enough to sustain a hostile work environment claim. The court said the conduct was offensive but not offensive enough.^[17]Rivera is not an outlier. In particular, one court noted that because emails are not open and pervasive, the hostility of the conduct is suspect.^[18]

A “like” is very different in that regard because Facebook content and their corresponding “likes” are not private exchanges. Although access can be limited through privacy settings, it is often made available to the poster’s friends, friends of the “liking” user and if the comment was originally posted to a third person’s wall, may include that user’s friends. A post by one employee may be available to a significant portion of the workforce. A comment posted on the victim’s wall is instantaneously available to his friends and family. The same is rarely true for workplace emails that can be forwarded to others but is generally not seen by as wide of an audience or beyond the workplace. However, even with the potential to be more pervasive than an email, it is unlikely Facebook conduct will be perceived by courts has severe or pervasive where a single “liked” comment is an isolated incident.^[19]

Where the Facebook conduct is combined with other harassment, the ultimate question is whether the “like” is part of a hostile work environment. Although the underlying “liked” comment may be analogous to verbal or emailed comments, the “like” merely expresses assent. While courts have included similar activity, such as co-workers laughing at offensive conduct, within the discussion of the plaintiff’s evidence of a hostile work environment the focus of the  analysis typically rests on the offensive conduct rather than the assent of others.^[20]

However, a “like” may be less like co-workers laughing at the conduct, passively assenting, and more like an actively sharing in the conduct by engaging in a conversation. Even in similar cases where a brief conversation with minimal, facially neutral comments by the second speaker courts are often identified as isolated incidents absent additional incidents. In Clark County School District v. Breeden, the Supreme Court ruled on such a situation. In Breeden, a female employee met with her male supervisor and a male co-worker. The supervisor read a sexual comment from a job applicant’s psychological evaluation and expressed a lack of understanding the comment. The male co-worker said he would explain later and they both laughed. The Supreme Court held it was an isolated incident and not severe enough to create a hostile work environment.^[21] Notwithstanding the public nature of Facebook, the same analysis can apply to a Facebook comment and “like”.

Continuing to assume the “like” occurs alongside other instances of harassment, the significance of the “like” becomes more important. Although the significance of a “like” has not been addressed specifically in an employment discrimination setting, a “like” has been addressed in other labor and employment law areas. In a 2012 case addressing “liking” a Facebook page in a public employment setting under constitutional protections, the Eastern District of Virginia delineated between Facebook comments and a “like”. In Bland v. Roberts, a sheriff’s civilian employee “liked” the page of the sheriff’s opponent in an election. After the sheriff learned of the support, he called a meeting and told his employees they should support him. After he won the election he discharged three civilian employees, including the employee who “liked” the opponent’s page, and replaced them with deputies. ^[22]

Disposing the plaintiff’s First Amendment claim, the court distinguished the substantive speech of a Facebook comment from a “like”. The court held “liking” was not a substantive statement warranting constitutional protections, unlike a comment with an actual statement where a definitive meaning can be derived from the words posted. The court declined to extend the same value to a “like” where no definitive meaning can be distilled.^[23] Although the constitutional protections analysis in Bland is inapplicable, the distinction in substance is applicable. Under Bland, a “liked” discriminatory comment may be some evidence of a hostile work environment but the “like” itself is likely not harassment by the “liking” user.

Taking the opposing approach, an administrative law judge in Three D addressed a Facebook “like” under recent National Labor Relations Board (NLRB) guidance, now formally adopted by the NLRB, regarding social media and worker rights under section seven of the National Labor Relations Act (NLRA).^[24] NLRA section seven protects workers who engage in concerted activity regarding working conditions.^[25] In October’s Costco decision, the NLRB formally adopted the position social media discussions between workers is protected concerted activity when multiple employees come together to discuss or protest working conditions. Under this policy, employer social media policies cannot be drafted in a way reasonably construed by employees as a prohibition on protected concerted activity.^[26]

Although the case is before the NLRB at this time, the administrative law judge presiding over Three D ruled a Facebook “like” in the context of employees discussing working conditions was protected concerted activity. In Three D, employees Sanzone and Spinella engaged in a Facebook discussion on Spinella’s Facebook wall with a former employee and customers regarding a dispute over tax withholdings with their employer, Three D-owned Triple Play. Spinella incited the conversation by posting, “Maybe someone should do the owners of Triple Play a favor and buy it from them. They can’t even do the tax paperwork correctly!!! Now I OWE money…Wtf!!!!” (emphasis in original). A few comments were made by non-employees. Sanzone “liked” Spinella’s comment. Spinella subsequently “liked” a separate comment from a former employee on that person’s Facebook wall. The next day Sanzone was fired for “liking” the comment. Spinella was fired the next day for both the comment and “like”.

Analyzing whether the “like” was protected concerted activity, the judge held it was sufficiently material to protect. The judge held the “like” assented to the comments and under the NLRA, concerted activity only requires a speaker and listener in a conversation related to the employment relationship. Additionally, the judge contended the Facebook conversation, including the “likes”, was an extension of an ongoing effort to correct the withholding issue with management.^[27]

Applying Three D to a hostile work environment analysis requires little manipulation, specifically where the Facebook comment and “like” are an extension of other workplace hostility. Judge Esposito’s analysis in Three D recognizes the “like” is by itself relevant as assent to the comment “liked” as part of a larger ongoing conversation between management and employees.^[28] While not required for a hostile work environment allegation, courts could easily draw the line of workplace conduct under Title VII where Facebook activity relates back to activity within the physical workplace, particularly when the Facebook comment and “like” are made within the workplace during work hours. Three D not only speaks to whether the “like” is a meaningful expression but may also assist in the earlier analysis defining workplace activity in a hostile work environment claim.

Three Dtakes the opposing view from Bland of the significance of “liking” content. Three D may more likely influence courts in discrimination cases for two reasons. First, the standard for harassing conduct in a hostile work environment claim is more similar to the standards under the NLRA than the more stringent constitutional requirements. Second, courts frequently look to NLRB decisions to guide employment discrimination decisions making it the more probable inspiration.^[29] Although prior consideration of NLRB decisions is not alone justification to forecast agreement with Three D, it is not improbable that courts would overlook a decision on-point factually and analytically.

Undoubtedly the issue is far from resolved. Although Three D would require a broader interpretation of workplace boundaries it would fairly balance the employer’s duty to correct workplace discrimination and recognize the limited ability to regulate conduct beyond its physical boundaries. Nonetheless, courts may continue to define the workplace narrowly and deemphasize electronically transmitted offensive conduct even if that rule requires ignoring the realities of how social media permeates the physical borders of the workplace.

III.             What’s an Employer to do?

Ultimately, the judicial guidance on Facebook “likes” and similar social media functions will have to rectify how hostile work environments arise in the modern context of a workplace rarely bound by the physical dimensions of the workplace. Beyond the mere “like” there are larger issues unaddressed with the modern workplace with cloud computing, remote workspaces and employees with unfettered access to the internet in the workplace. Although employers have dealt with similar issues with phones, email and texting in the workplace, social media presents a different problem due to its public nature and ability to simultaneously implicate workplace activity and off duty conduct.

Neither the conservative nor liberal approach to defining the workplace provides a viable standard for resolving these social media questions.The conservative approach leaves too many questions unanswered. It is a functional approach when the off duty harassment is something that cannot bleed into the workplace but social media-based harassment is clearly different. The more liberal approach accepting off duty harassment as part of a hostile work environment would allow relevant evidence to be presented but also risks being too open-ended. Some social media activity may be offensive but too remote to the workplace to be directed at another employee or part of a hostile work environment. A “like” could just as easily apply to an offensive comment but wholly unrelated to workplace activity as it could be two co-workers intending to harass a co-worker through Facebook.

 Both Blakely and Amira-Jabbar deficiently address the multitude of ways social media can be employed to harass. Blakely provides a reasonable method to address  whether liability is imputed on the employer when employees use its social media outlets to harass but in an environment where employees can harass each other through private Facebook accounts in the workplace Blakely’s substantial benefits test may not effectively regulate all social media-based harassment in the workplace. Similarly, Amira-Jabbar fails to provide a functional rule in addressing how social media-based harassment infiltrates the workplace. Blocking access to social media on employer devices deals with the symptoms of the problem – avoiding seeing the harassment rather than preventing or correcting it – but leaves the harassment to continue, even within the workplace on personal devices. Even a synthesis of these two rules leaves open too much social media activity and too many unresolved questions to viably resolve the issue.

Instead, the more flexible rule fairly balancing the employer’s duty to prevent workplace harassment and the employee’s desire to be free from it is adopting Three D in a hostile work environment analysis. Under this adaptation the physical workplace remains the focus of the inquiry. Social media-based harassment within the physical workplace, whether on employer or personal devices would be harassment because it is within the physical workplace. Recognizing the modern workplace is not always a stationary, employer-controlled premises, the same rule would apply to any place an employee is performing work. This is no more expansive than harassment by any other medium. An employee working from home could call a co-worker at the employer’s worksite and harass over the phone, equally as offensive as a call placed between two offices in the same building. The same should be true for social media.

However, when social media-based harassment occurs beyond the workplace, Three D provides a reasonable approach to distinguishing work-related and remote, independent behavior. When the off duty activity relates back to workplace activity, it is an extension of the harassment in the workplace and should be dealt with as part of the same problem. As the Blakely court recognized, if the employer does not deal with off duty but work-related harassment it sends the message that work-related harassment is acceptable. To prevent exacerbating harassment within the workplace the employer should be responsible for correcting work-related harassment occurring off duty when the offended employee brings it to the employer’s attention. The offended employee has the burden to prove the relationship between harassment in the physical workplace and the social media-based harassment.

Similarly, Three D provides a standard to assess social media activity at work to distinguish harassing activity from merely offensive activity. An employee may “like” content another employee finds offensive but unless the “liked” content specifically relates to the offended employee and other evidence of harassment in the workplace it would not be evidence of a hostile work environment. Offensive conduct on Facebook would not be enough without the relation to other workplace harassment.

            The exception lies where the social media activity is so offensive by itself, although an isolated incident, is enough to give rise to a hostile work environment. Given the reluctance of courts to find an isolated incident sufficient evidence of a hostile work environment, it is unlikely any single posting or “like” can meet that burden. However, it is not impossible the content could be so offensive that combined with the public exposure it could meet that burden.

Under Three D, a Facebook “like” could be evidence of a hostile work environment. The Bland court held the “like” was not a meaningful expression; however, a hostile work environment analysis looks at the effects of the expression, not its meaning to the perpetrator. Certainly in a hostile work environment claim the underlying content “liked” must be harassment but the “like” elevates it. The “like” shows assent by others in the workplace, suggesting the harassing view is shared and will likely encourage more harassment.

Additionally, the “like” further publicizes the harassment. When content is “liked” Facebook shares it by adding a note on the “liking” user’s wall that the comment was “liked”. That user’s Facebook friends can then view what was liked. For that reason, it is more than just passive assent, it actively spreads the harassment. Under the totality of circumstances test applied to hostile work environment claims, the “like” fuels several factors. It fulfills the factor observing that multiple employees joined in the conduct. It can also aggravate the humiliating and offensive nature of the underlying content by making it more widely exposed, especially if the harassment is posted to the victim’s wall where her friends and family can see it.

This rule does not fundamentally alter precedent assessing the weight and value of Facebook activity. Courts can still assess social media as less offensive than face-to-face encounters, as they often do for emails, while recognizing the public nature may make it more offensive than an email. Additionally, courts can continue view individual events as minor but  included within the events comprising a hostile work environment. The admission of social media activity does not mean every click or word typed becomes a potential hostile work environment.

Three Dprovides reasonable limits to social media activity relating to hostile work environments by balancing the policy interests in Title VII and other anti-discrimination laws. It preserves the traditional boundaries of the workplace while recognizing work-related activity does not always end at the front gate. Under the current hostile work environment case law a “like” is unlikely to provide much probative evidence for a plaintiff but as more claims come forward alleging social media-based harassment, judicial attitudes may change to reflect the realities of the changing workplace. Employers should be mindful that courts will receive invitations from plaintiffs to address the questions raised here. For that reason, employers should contemplate how their social media policies will deal with those issues and how to handle the employee that complains about co-workers “liking” harassment.