Michael Ledford: Confession Falsified by Space and Time

For anyone attempting to understand a crime, or an alleged crime, a comprehensive and accurate timeline is a thing of beauty. Even our rapidly advancing technology does not allow a person to be in two different places at the same time. Even the most sophisticated forensic analysis cannot cause time to run backwards. Even the most rabid or obfuscatory closing arguments cannot erase the impact of evolving testimony or altered documents.

Useful as they are, however, comprehensive and accurate timelines can be challenging to construct. A flimsy timeline constructed of surmise and unsubstantiated claims is of considerably less value than a robust timeline constructed of irrefutable facts. Good luck building the robust version.

In this post, I'll attempt to present a timeline of events for the fire that occurred at Michael Ledford's apartment in Stuarts Draft, Virginia on the tenth of October, 1999. The fire resulted in the asphyxiation death of one-year-old Zachary Ledford (Michael's son), the severe burning of Elise Ledford (Michael's wife), and the fifty-year prison sentence in Virginia (a state with no parole) of which Michael Ledford has now served a decade.

Keep in mind as we work through the mundane events, dates, and time that we are actually attempting to determine whether Michael Ledford is indeed guilty of murdering his one-year-old son and seriously burning his wife, or whether he was wrongfully convicted of a horrific crime. While the tasks may seem mundane, the objective is, quite literally, truth and justice.

The Birthday Party

Most of the events of that day and that evening are undisputed. Michael and Elise Ledford held a small birthday party for their son Zachary who had been born one year and three days earlier. Guests included Elise's parents and two family friends.

According to Elise's testimony, after the guests left she cleaned up, gave Zachary a bath, and gave him a bottle. She put him in his crib (in his own room) sometime between 8:00 and 8:30 PM, after his hair had time to dry. Soon thereafter, she went to bed herself, since she had been up since 4:00 AM and since she was scheduled to be at work early the next day. She remembers nothing else until being roused from a medically sustained coma five weeks later.

According to Michael's testimony, he arose sometime between 8:00 and 8:30 AM. He tended and played with Zachary while Elise worked her morning shift. Elise returned home around 11:30 AM. The party broke up between 3:30 and 4:00 PM. Michael laid Zachary down for a later-than-usual nap while Elise spoke outside with her parents and the family friends.

Michael's mother, Pat, called from her home in Pennsylvania, as she usually did on Sunday afternoons. She asked about Zachary and the party. Michael assured her that Zachary liked the gift she had bought for him.

Once the guests had left, and the baby was napping, and the phone was quiet, Michael and Elise made love for what would be the last time.

Soon thereafter, Zachary awoke. Either he or Elise fed Zachary while the other finished cleaning up after the party. They both got down on the floor and played with their son.

Michael's Departure

Elise put Zachary to bed around 8:00 PM and got ready for bed herself. She asked Michael if he would put some gas in the car. He said he would and that he would stop just briefly at the fire department to sign his training log. Michael, as it turns out, was a firefighter for the Stuarts Draft Volunteer Fire Department.

Michael estimates that Elise went to bed sometime between 8:00 and 8:15 PM. He watched TV briefly, in the dark, for money was scarce and they were careful with their utilities. He left to complete his two errands sometime between 8:30 and 8:40 PM. Michael's neighbor (the one in the apartment across the entry way) reported hearing Michael leave (Michael's car has a distinctive loud noise) approximately 30 minutes before the fire trucks arrived. Since they arrived at 8:57 PM, Michael's neighbor estimated that Michael left around 8:27 PM. That is quite close to the earlier end of Michael's estimate.

As Michael was leaving, he turned off the television then flipped on the light switch near the entry door, as a courtesy for their boarder who would be arriving home soon.

It is at that moment Michael was leaving that the only substantive disagreement occurs. The entire case hinges on what Michael did or didn't do as he left that evening. According to his testimony, he did nothing other than flip the light switch to the on position and shut the door behind him. According to the State of Virginia, and according Michael's recanted confession, he took a long and slender candle from a pair that was burning on the end table, lit it with a cigarette lighter, and tossed the candle into the chair before leaving and shutting the door behind him.

The Gas Station

No one disputes that Michael then drove from 19 Highland Drive to Little's Exxon (now a Citgo) station on the corner of Stuarts Draft Highway (U.S. Highway 340) and Draft Avenue (State Route 608). Zachary's babysitter coincidentally drove past the gas station at approximately 8:35 PM and noticed the Ledford's burgundy/maroon Cavalier there. The car was distinctive due to its red rectangular-shaped light on the roof.

Given that Google gives a driving distance of 0.3 miles and a driving time of 2 minutes, the babysitter's time is consistent with Michael leaving near 8:30 PM. It is not consistent, however, with Michael leaving near 8:40 PM. The babysitter could, of course, have been off in her estimate.

Michael put $9 or $10 worth of gas in the car. In April of that year, gas prices in Virginia were around $1.03 per gallon for unleaded, according to my online search. Michael apparently pumped somewhat more than 9 gallons of gas into his car. He went inside to pay for it and spoke briefly with the cashier. I'm unaware of any effort by the police or the defense to interview the cashier, recover receipts, or view any security tape.

In any case, it would take a finite amount of time to pump and pay for the gasoline. The Skeptical Spouse and I each guessed it would take 5 minutes. I then recreated the transaction as best I could, using my vehicle and my local gas station. I started my stopwatch as I pulled into the gas station. I parked without delay at an available pump. I walked inside and prepaid without delay. I pumped 9.3 gallons of gas into the my car. The pump seemed to be pumping at a normal speed, neither noticeably faster nor slower than expected. I returned the pump handle, reentered the car, and left without delay. I stopped the stopwatch as I pulled out of the gas station. The time was 5 minutes and 2 seconds.

Assuming the babystitter's time of 8:35 was correct, and assuming my 5 minute transaction estimate is correct, Michael arrived at the station sometime between 8:30 and 8:35 PM. He left sometime between 8:35 and 8:40 PM.

The Fire Rescue Station

From the gas station, Michael drove along Draft Avenue either "to the Stuarts Draft Rescue or the Stuarts Draft Fire Department." The two buildings are across the street from one another, so it makes little difference from a timeline perspective which he went to, though I would like clarification. For now, I'll assume he went to the Stuarts Draft Rescue building, which is slightly closer and much easier to locate on Google. (Google whiffs when attempting to locate 118 Draft Avenue.)

Google estimates the distance to the rescue building to be 0.2 miles, and estimates the driving time as 54 seconds. We'll call that one minute. Assuming Michael left the gas station no later than 8:40, he arrived at the rescue building no later than 8:41. Given all the other assumptions already described, Michael arrived at the rescue building between 8:36 and 8:41 PM.

Tenants Discover the Fire

Slightly later, at approximately 8:45 PM, Deborah Moore and Jim Dorsey became aware of a fire in one of the Highland Hills apartments. They had just finished moving into a nearby unit.

"As we were getting ready to pack up the tarps and everything that we had used to bring everything down on a trailer ... I heard ... glass pop; a popping sound. And when I looked to the window, I saw flames kicking up from the bottom right side of the window; the large windows. ... I couldn't tell what was making the flames, but within seconds, the drapery went up and the whole window was engulfed in flames. ... For a few seconds, my boyfriend and myself stood on the sidewalk, just watching the window. And then we heard a woman scream inside. I instantly ran up the stairwell into the entryway; to the door; tried the handle; tried to open the door. The handle was cold; it wasn't hot. It was locked. I banged on the door and told her to open the door. My boyfriend was yelling to tell her to go to the back windows to get out. And so I continued banging on the door, telling her to go to the back. And then he yelled for me to get out of the entry hall, in case there was, you know, an explosion with gas or something. ... I thought I had heard a minimal whimper ... just ... a whine, or something like that. But I didn't hear anything else."

Jim Dorsey recalls seeing Michael leave 20 minutes (or less) prior to the time he noticed the fire. Since he noticed the fire around 8:45 PM, that places Michael's time of departure at 8:25 PM or later. Once again, an 8:30 departure seems reasonable. An 8:40 departure seems quite late, since that would have only been 5 minutes (not 20 minutes) before Jim Dorsey noticed the fire.

911

At 8:47 PM two people called 911. One of them was a neighbor. One of them was Elise Ledford. She wasn't able to say anything. The operator only heard her gasp, then the line went dead.

At 8:50 PM, a 911 operator called the Stuarts Draft Fire Department and informed them of a fire at the Highland Hills apartment complex.

Michael's Return

Michael had finished recording his training on a log inside the fire rescue building when tones activated on his pager. The pager indicated a possible structure fire at the Highland Hills apartments complex, the complex where he lived. Highland Hills is a large complex, though, and it did not occur to Michael that  it might be his apartment that was on fire.

"I thought, you know, I, I live up there. I didn't see anything when I left. Someone probably just panicked, probably burned something in the oven, so instead of waiting to ride in a fire truck, I just went up in my personal vehicle."

Michael did not endanger others as he returned to the complex. In fact, he stopped at the red light at the intersection near the gas station where he had just fueled his car. Google lists the distance back to the apartment complex as 0.5 miles. It lists the driving time as 4 minutes. I don't know how Google accounts for the variations that might be caused by red lights.

As he approached the apartment complex, Michael slowed and looked down one of the two long rows of apartments. He saw nothing. An excited woman, who turned about to be the aforementioned Deborah Moore, noticed the red light on the top of his car, flagged him down, and informed him that the fire was at the top of the hill in the next row of apartments. Michael drove directly up there, found he had been preceded by another volunteer firefighter (who lived in the complex), and realized only then to his horror that it was his apartment that was on fire.

He rushed to the back of the apartment to see if Elise had escaped or attempted to escape out a window. As a firefighter, Michael had been concerned about the single exit from the building; he had advised Elise that she could escape through the rear windows should a fire ever block the exit.

The rear windows were closed and darkened by smoke. The prosecution would attempt to cast doubt on Michael's story by questioning why he did not break the rear windows and attempt a rescue of his wife and son. Michael explained he did not do so because the windows were too high and because his training told him not to do so. He did not explain, nor did anyone else for the defense, that he was ill equipped to do so, would likely have succumbed quickly to the smoke and toxic gases, and would  have therefore been unlikely to succeed. Nor did anyone explain that had he attempted such a rescue, it is possible that Elise would have died along with Zachary. The fresh air would have stoked the fire which was in the process of self-extinguishing due to lack of oxygen. The fire training he received took into account all such factors.

Michael returned to the front of apartment where he had to be physically restrained by the properly-equipped firefighters just arriving. The first fire trucks arrived at 8:57 PM. Assuming Michael heard the tones at 8:50 and spent 4 minutes to return to the complex, that places Michael at the apartment 3 minutes before the first fire vehicle arrived. That seems consistent with the evidence and testimony.

Elise's Response to the Smoke Detector

Of particular interest to us is when the fired started relative to when Michael left the apartment. To make that determination, we will need some additional information which is thankfully available. Assuming Elise was awakened by the smoke alarm, we will need to know how long it takes for a smoke alarm to awaken sleeping occupants, and we will need to know how long the fire had burned before it triggered the smoke alarm.

I located online a number of studies investigating the effectiveness of high-pitched smoke alarms in waking occupants. The controlled studies did not involve actual fires. I can summarize the results pretty quickly.

Young children will not awaken to high-pitched alarms. They awaken most consistently to female voices. Some organizations have therefore recommended altering or supplementing the type of alarms emitted by smoke detectors.

People with hearing difficulties are obviously insensitive to smoke alarms. Seniors fall disproportionately into this category. This rather obvious finding has also led to recommendations for supplemented or altered alert sounds.

Sleep deprived college students sometimes fail to awake to high-pitched alarms.

All adults without hearing impairment awaken within 30 seconds when subjected to the high-pitched tone associated with smoke alarms.

I can provide the results of a small study based on a single, anecdotal event. When I was twenty-nine years old, the smoke detector in our house went off at 4 AM. I not only awoke instantly, I believed I had set a world's record from fully asleep to standing beside the bed searching for flames licking at the door. My time estimate for that sequence is 7 nanoseconds. I dismiss the possibility that I slept for 30 seconds and then awoke rapidly because my wife tied my 7 nanosecond record from sleeping to standing. It's statistically unlikely we slept for the identical 30 second, or 15 second, or 5 second interval before initiating our record-setting 7 nanosecond feat.

In our case, the alarm turned out to be a false. After checking the house for fire, I insured all gas sources were turned off and I popped all the circuit breakers. I removed the battery from the stupid fire alarm and returned to bed. There we joked and giggled nervously until it was time to get up.

Perhaps not coincidentally, the fire experts paid by Michael's defense team assumed that Elise awakened thirty seconds after the fire alarm sounded. I will disagree only slightly. I will assume that Elise awakened within thirty seconds of the alarm sounding, assuming it sounded at all.

The Smoke Detector's Response to the Fire

The Court allowed Michael's defense team to hire fire experts. The defense hired the Maryland firm Combustion Science and Engineering. CSE received a $15,000 fixed fee to research, report, and testify about the case. I offer the company's mission statement directly from their web site.

For more than twelve years, Combustion Science & Engineering, Inc. (CSE) has been dedicated to the study, advancement, and application of combustion and fire science. Combining a wealth of knowledge and experience, from the private to public sector, from academia to industry, CSE’s spirited partnership and dedicated team offers exceptional technical leadership, intelligent solutions in combustion and fire protection, and superior fire and explosion investigations.

Since the fire occurred 11 years ago, it seems CSE had not been in business for a long time before being called on to assist in the defense of Michael Ledford.

Steven Olenick (master of science in fire protection engineering) was the principal author of the company's report. Dr. Richard Roby (founder and principal of the company) testified at the trial. Together they prepared and reported on (among other matters) the speed with which the fire burned and spread. Using the best modeling technique available at the time, one developed by the National Institute of Standards and Technology, CSE estimated that fire started between 8:43 and 8:45 PM.

The jury was unimpressed by the testimony of hired experts, believed Michael when he confessed, didn't believe him when he testified, and voted him guilty.

CSE seemingly never lost interest in the case. In 2010, on their own dime, they prepared and published a paper entitled "Re-Visiting the Michael Ledford Fire Incident." In that paper they discussed the results of their re-evaluation of the Ledford fire using the latest fire modeling computer simulation technique. I have read the paper carefully, am familiar with finite element modeling (since we used it to model the aerodynamic behavior of aircraft) and I am dutifully impressed by their work. I include below a particularly relevant segment from their latest report, and include the corresponding image. Click on the image to enlarge.

In the model, the smoke alarm activated at approximately 75s. Adding in 30s to awaken to the alarm and 30s to respond, it is expected that Elise would be in the living at approximately 135s after the initiation of the fire. At 135s, the approximate time she would be expected to be in the living room under the confession scenario, the fire size is approximately 100 kW. This fire size would likely be perceived as small enough to approach and attempt to mitigate, but large enough that she would be unable to do so. Additionally, the smoke layer would be low enough to cause Elise distress, while remaining high enough so that the witnesses would be able to see her in the living room. A graphic showing the fire size as well as a graphic showing what the witnesses would see in the living room from a vantage point outside the residence at the time Elise would be expected to be present are presented below as Figure 8. The size of the fire and smoke layer in the model at the time Elise would be in the living room is in agreement with the witness accounts of the fire when they saw Elise through the window.

It is a terrific piece of work. If interested you can read the entire paper from Pat Ledford's website, Innocent in Prison. If you are satisfied that I am reporting accurately, you can simply accept that CSE estimated that smoke detector triggered 75 seconds after the fire began. They assumed that Elise arose 30 seconds later, and placed her in the living room, witnessed by Deborah Moore, 30 seconds after that. For reasons not made clear in their paper, CSE estimates Elise would not call 911 until more than 3 minutes after being witnessed in the living room. I find that time estimate to be surprisingly large and I respectfully disagree.

I assume CSE's 3 minute time delay results from their insight into how much smoke was generated and how much smoke one can breath before becoming unconscious. I note, however, that it is only an assumption that Elise passed out immediately after the 911 operator heard her gasp. The 911 operator in fact testified that the line went dead. I believe, based on the photo of the electrical service panel that I will reserve for a later post, that the line went dead because the circuit breaker popped. Elise's portable phone required continuous power to the docking station located in the kitchen.

Finally, The Fire Start Time

Instead of 3 minutes, I assume Elise called 911 soon after discovering the front door blocked by fire, that she called within 45 seconds of realizing she was trapped. That gives a total time of from the start of the fire to the 911 call of 3 minutes. That three minutes consists of 75 seconds for the fire to trigger the smoke alarm, 30 seconds for Elise to awake because of the alarm, 30 seconds to make her way from the bedroom to the living room, and 45 seconds to call 911 after realizing she was trapped. Each of the last three increments seems conservative.

Given that the 911 call is well documented as having taken place at 8:47 PM, and given that I believe the evidence indicates Elise called within 3 minutes of the start of the fire, I claim the fire started no earlier than 8:44 PM. Given that Michael left the apartment no later than 8:40 PM, and more likely left the apartment around 8:30 PM, I conclude (as did CSE before me) that Michael could not have started the fire.

More Testing Required

I recognize that someone convinced of Michael's guilt could reasonably argue that the fire smoldered for 4 or 14 minutes before spreading rapidly. I discussed this point briefly with CSE via email. They believe this to be unlikely, and they explain that their model takes ignition time into account. They do, however, acknowledge that others may not accept their modeling of this issue. To further address the issue of ignition time, Pat Ledford, fire investigator Brian Hattem, and I intend to conduct testing that will consist of repeatedly tossing candles into chair analogs, chair-like objects. We promise to be careful.

If the candles never set fire to the chair, then the testing will add to the argument that Michael's confession is falsified by space and time. If the candles set fire to the chair almost immediately, once again Michael's confession would be falsified.

Only if at least one of the tests resulted in an extended smoldering before a rapid fire spread would Michael's confession withstand the rigors of a detailed timeline analysis. That would not mean that Michael's confession was true, since it is already falsified by thermodynamics. It would mean, however, that the challenge before us is even harder than we had anticipated.