The pen was anchored by a series of ship anchors, several tons each, that ran off in many directions. They were not visible from the surface, but later in the project, I would get a good look during maintenance dives. The first impression is that of a completely tangled mess, but they were, in fact, a systematically balanced and tensioned mathematical wonder that kept the pen from complete annihilation in pretty insane weather and currents. The operations team constantly worked to achieve the ideal tension equilibrium amassed between the maze of cables. If one side or line was off by just enough, the resulting imbalance could swiftly break the pen to pieces.
A very course grid work made of fiberglass called “Chemgrate” was laid horizontally about a foot or so above the structural tubes. This constituted the deck and made the bay pen walkable. The surface coating had to be super rough in order to provide a stable foot-grip, but if any ever fell on it, they might fare better dragging their face across a cheese grater. Outside of the main deck areas in the middle of the pen, the grate only provided about a two-foot-wide passage around the expanse of the two main pools. Exterior handrails made of the same high-density plastic kept us from being blown off the deck and into the bay, but nothing offered protection from taking a plunge into Keiko’s side of the pen.
Underwater, nets hung from all sides beneath the structural tubes that formed the shape of Keiko’s pools. They completed the pen’s confinement perimeter. The bottom of the facility, about thirty feet deep, was also netting, but attached to the vertical net walls by a large concrete ring underneath, constituting the entire diameter of each of the two main pools. The rings weighted the net, maintaining the pool’s shape and providing somewhat of a sea anchor to the structural integrity of the pen itself. At low tide, the suspended bottom of the bay pen was only a couple feet from touching the ocean floor. During high tide, it might extend as much as twelve to fifteen feet from the bay’s floor, stretching the anchor cables to their fullest. There were times that this extensive variation occurred in minutes rather than hours at the hands of many violent storm surges that plagued Klettsvik Bay in the winter months. The befuddling matrix that formed the anchor system was actually the front line in the bay pen’s survival.
The “research shack” was out of this world. Never in my wildest dreams did I expect to find such plush accommodations on the pen. Heck, I was even amazed that there was power, or at least hardwired power! I expected things to be run from a generator. Nothing doing, and although they had a backup generator the size of a Volkswagen behind the dive locker, the main power and a phone line were run along the bottom of the bay straight out from the town. They even had Internet access!
The light green research shack was like something found on a construction site, only this one was in much better shape. Measuring approximately thirty feet long by ten feet wide, there were only two doors to enter or exit: one on the end and one roughly in the middle of its length. There were windows, too, but only at the southern end providing views of Keiko’s pool, the harbor to the southwest, and the cliffs to the east.
We entered through the door in the end and walked into a bunking room that doubled as a “wet room,” an area for disrobing survival suits after the routine dousing from the elements. From there a small foyer and bathroom joined the wet room to the “dry room” where the staff spent a majority of its time. The dry room had a small kitchenette with sink, running water, a coffee maker, microwave and cabinets full of more dishes and kitchenware than I had in my first apartment. At the back end of the dry room was a bank of video screens, about nineteen in total, providing images from all around the pen, including a few from underwater. There was also audio recording equipment and a few hydrophone hookups that allowed the staff to listen to or record underwater sounds through submerged microphones.
One of the hydrophones was connected to a speaker, always providing the constant low-level underwater sounds, echo-like bumps and grinds of the ever moving bay pen. Under the window facing Keiko’s pool, a low counter provided desk space and included shelves above. There was a lone computer for staff use on the pen sitting beneath the east-facing window looking out toward Keiko and the interior of the bay pen. There were even blue and white flowered Midwest-style curtains. Three cafeteria-like chairs completed the accommodations.
Very cool, I smiled. As a person from the animal field, I was not used to having all these work-related toys. After a brief tour of the bay pen housings, including an explanation of the records taken on Keiko, ethogram data recorded on Keiko’s activities, and various other procedures and protocols, we went out on deck to watch a training session with Keiko, the “Big Man,” as I would come to call him.
Thrashing
Stephen Claussen slapped the water’s surface, the signal for calling Keiko over to where he stood at the pool’s edge. Stephen was the lead trainer on this particular staff rotation. Stephen had gained his whale experience caring for Keiko in Oregon. He was full of nervous quirks. At times Stephen would unknowingly rub his hands together, one balled inside the other as if the evildoer in a cartoon escapade. Other times he would do it consciously, acting out the backdrop of a twisted comment. He was an immensely funny guy. His sense of humor was often a great and welcome equalizer in the middle of our newness, dampening the uncertainty pressed upon the staff. Stephen and I became fast friends.
The session was painful to watch. I had never seen such a slow whale. It was as if I was watching a fully loaded dump truck double-clutch through thirteen gears to get moving. Keiko, when he finally came over to Stephen, didn’t even lift his eyes above the waterline. This posture is analogous to a person who “just-woke-up” dazed and with his or her eyes half shut. Hello? Are you hearing me? One can never be sure.
Stephen stood slightly hunched over, his chin almost on his chest as he peered down at Keiko. He nervously talked to him, his whistle bridge clenched between his teeth, narrating the more obvious while he pondered his next steps. (A “bridge” is an audible whistle signal that “bridges” the gap between the completion of a correct behavior and the whale receiving reinforcement.) Stephen’s posture didn’t lend much to a professional appearance. Instead, the way he carried himself made his clothes, the same apparel most of us wore, appear on him just a bit more disheveled.
Stephen moved ahead with his session plan, asking Keiko for a few behaviors. Among the menu of trials he gave the signal for a behavior they called an “innovative.” Having no idea what I was watching, the session seemed to drag on with no end in sight. I was like a five-year-old sitting in church, fidgeting and struggling to pay attention. An eternity seemed to pass before the session was finally over. At last, Stephen stood upright, gave yet another signal with both hands, then promptly walked away from the session.
Afterwards I walked up on the bridge that connected the two main decks of the bay pen, the best vantage point to observe both the north and south pools. Keiko stayed right where they ended the session for a minute or so, and then turned slowly on his side violently throwing his head down, mouth open all the while and shaking one massive pectoral flipper wildly in the air. Rolling back to an upright position he took one explosive breath and started a slow counter clockwise swim in the north pool. What the hell was that all about? I wondered.
The odd behavior I had witnessed after the session ended was referred to as “thrashing” by the staff. To me, it appeared like frustration, a temper tantrum. Apparently, this was something that occurred with regularity and not necessarily during or immediately after training sessions. Sometimes it occurred seemingly at random, but always during normal daytime work hours. To my knowledge, the thrashing behavior never occurred at night or when trainers were ordinarily not present.
Standing atop the bridge I pondered every possible factor that might surround this odd behavior. I stared down at Keiko as he returned to his presession spot and slowly drifted to a stop at the surface. The sight of him broke my train of thought. If I didn’t know better, I would have thought this was a preg
nant female. I had seen fat dolphins before, but never a fat killer whale. From directly above, he looked like a giant guppy with his dorsal muscle ridge framed against a bulging undercarriage. Wow! I wanted to talk to Robin badly, but with the staff around and not wanting to insult my new acquaintances, there wasn’t much I could say, at least not yet.
Ground Zero
Where I came from, killer whales were wickedly alert, fast and bright-eyed. In the training environment, a trainer gets what he or she pays for … meaning that behavior follows reinforcement. Further, dominant traits of an animal in a training scenario often betray the tendencies of the animal’s trainer or trainers. They are clues to where the majority of effort has been focused (sometimes with purpose and sometimes completely unwittingly).
Think of a dog that jumps on its owner every day when the owner arrives home. The owner says “no” in a stern voice. Nevertheless, without fail, the dog continues jumping on its owner each day. Simply put, the owner is a reinforcing quality in the dog’s life. Regardless of what the owner says when the dog jumps, the dog is getting attention from its owner, so the behavior is strengthened.
In this context, the word “reinforcement” refers to what happens immediately “after” a behavior to strengthen or increase the likelihood that the behavior will occur again. Not all consequences are reinforcing. Punishment for instance, has the opposite effect, reducing the behavior it follows. Specifically, reinforcement is said to “reinforce,” strengthening the specific behavior it follows. If the consequence of a behavior causes that behavior to increase in frequency then that outcome is reinforcing (empirically defined). More to the point on Keiko, if “slowness” is reinforced, the result is a lethargic animal. On the other hand, if an atmosphere of high energy is cultivated by consistently reinforcing quick responses and attentive behavior, the result is a highly engaged and responsive animal.
“Sd” is the designation for Discriminative Stimulus. It is a specific signal that requests a specific response or behavior from the animal. Sd’s can be hand signals, like sign language, or they can be audible signals or even environmental cues, like opening a door or turning on lights. Keiko did not know any audible or other forms of Sd. The vast majority of his learned repertoire was based on hand signals.
Behavior is a science; the application of behavioral modification (or training) is equally exacting. Among the many tenants of this practice, a trainer should not provide a signal or Sd asking for a response that he cannot reinforce or does not intend to reinforce. At the end of training sessions with Keiko, the staff would signal to Keiko that they were finished. It was the same signal that I had seen Stephen use. By giving Keiko a signal that they were “breaking” from the session (or ending the session) Keiko’s caretakers presented an S-delta, a signal that indicates reinforcement is no longer available. In fact, theoretically, the signal itself becomes negative because it communicates to Keiko, “I’m leaving you now and taking my reinforcement with me.” To Keiko’s trainers it was a courteous, simple communication. To Keiko, it set the stage for frustration. Think of a toddler when he first recognizes the cues that Mom is leaving.
Schedules can also lend themselves to aggravation. If a session schedule is so routine that it becomes predictable, added to a “breaking” signal (delta), frustration can, and eventually will, escalate to its close neighbor, aggression.
There were many signs that Keiko’s training stemmed from trainers with a background of pseudo-behavioral experience. These trainers, great of heart and talented in areas, never understood the science of behavioral modification, but rather, had techniques passed down to them through the school of hard knocks.
Another misguided construct involved that of the “innovative” behavior. This was a signal that was given to Keiko, and in response he could do (was supposed to do) whatever he wanted. The only requirement was that he could not do the same behavior twice in a row. Each time, he had to do something unique in order to receive reinforcement. They believed this was a fun interaction; that it stimulated Keiko to be creative and independent. In reality, it was yet another gray area of confusion for Keiko: no clear criteria, no clear direction and no consistency in the result. Spontaneity has its place in life, but not as a trained behavior. The two are mutually exclusive. The “innovative” was only the tip of the proverbial iceberg. In fact, most of Keiko’s training interactions were, to borrow from Douglas Adams, “somewhat similar to but totally unlike” behavioral modification, leaving in their wake a host of aberrant and self-destructive habits in Keiko.
Years ago, as an up-and-coming trainer in the SeaWorld system, I was requested to appear at the office of the curator of Animal Training, Thad Lacinak. That particular day, Thad was perturbed by some stupid mistakes made by trainers which he had witnessed during a show; mistakes that he saw as creating confusion in the whales. A trend of confusion quickly leads to frustration, a predecessor of aggression. Unless corrected, this pattern of mistakes would ultimately get someone hurt. As one of the few “waterwork” trainers (trainers approved to be in the water with the whales), he challenged me with a simple question: “What happens if you bridge a behavior, the animal ignores the bridge, and you bridge again?” He was talking specifically about killer whales. At six foot three, Thad was an imposing boss. Adding intensity to his question, his normally pensive expression was painted with stern urgency while he awaited an answer.
The whistle bridge asks the whale to stop what it is doing and return to the trainer for reinforcement. Killer whales are top predators. Their hearing, eyesight and sonar abilities combine to make them the most aware animals I have ever been around. If a whistle bridge is blown, there’s no such thing as “they didn’t hear it.” I responded to Thad’s question, “That’s a surefire way to get your ass chewed.” I had passed the test. Continuing to provide direction when that direction is being purposely ignored is one of the fastest ways to ignite frustration and aggression in any animal (or person), especially a killer whale.
Yet again, this was another sloppy area in Keiko’s interactions with his trainers. They would frequently blow their whistle, in effect, telling Keiko, “Great, that’s it … now, come back and I’ve got a great reward for you.” Often, when Keiko would not respond to the bridge, they would promptly bridge again, insisting that his hearing was bad. Highly unlikely; it was more probable that this “two or three bridge requirement” was the result of Keiko training (or ignoring) his trainers. Any other whale might well launch itself bodily out of the water, gaping mouth twisting to the side as if to grab the unsuspecting offender or at the least knocking them aside in a sweep of its head. Having witnessed this exact response to repeated bridging before, I can submit with confidence that once is enough to learn the lesson. I will never underestimate just how remarkably swift a large killer whale can be when driven to the point of frustration. If Keiko hadn’t been so satiated with food, or been so numbed by this practice throughout his learning history, he might have reacted to these situations like a normal male killer whale and left the trainer with knocking knees for an hour or two.
These examples of conflicting signals, along with many others, were circumstances that should have totally pissed him off, yet the lack of aggression or even precursors to aggressive behavior from Keiko revealed another discomforting trait of this whale charged with surviving the wild: it was as if he had been “dumbed down” or dulled to the point of complete apathy.
In large part, the driving force behind Keiko’s lethargy was not only poor training, it was compounded by diet. Because Dr. Cornell had mandated that a top priority for Keiko’s release was to fatten him up, Keiko was completely satiated with food. Keiko simply didn’t care whether food was offered or not. The only motivation to interact with his trainers was the stimulation they provided, and the break it offered him from an otherwise monotonous day, void of social contact with other whales, stimulating mental challenges or any other form of variability aside from changes in weather or current. Occasional
ly, when Keiko would not even care to come over, the staff would literally throw herring at him to make sure he got all his food for the day. Often, without moving, Keiko would just watch the herring sink to the bottom.
Imagine being so satiated after a Thanksgiving dinner that a nap is all that holds interest. Friends call and want to toss the Frisbee. Boredom begs for agreement, but motivation is stifled by an overbearing impulse to lie motionless, drifting in and out of consciousness. This was Keiko. This was the whale destined for release to a supremely unforgiving environment.
His fattened and lethargic condition had become a smoke screen that clouded any true evaluation of the animal. First, we would need to get Keiko on a workout regimen. We needed to get him moving and burning more calories, but not necessarily dropping his food intake in the process. Doing both at the same time might trigger the opposite result for which we were aiming. A drop in fuel at the same time we turned up his calorie burn could push his body to store more reserves, more fat. Robin and I couldn’t contain our need to discuss the issues and eventually allowed ourselves to share our observations openly in front of and with the staff. During the following weeks and months, we would find ourselves constantly educating the staff on the basics of behavioral modification, normal killer whale behavior, nutritional dynamics and physiology. But talk is cheap; we would need to show them results.
Killing Keiko Page 6