However real and significant synchronicities can seem, they are also notoriously subjective. The events of a synchronicity can be simultaneous or years apart. They can be strikingly similar, or one event can merely seem like the answer to the other. What seems beyond chance to one person can seem random to another. What looks like a clear message from one angle can be interpreted differently from another angle, or be viewed as altogether meaningless. How can we study a phenomenon that exists so fully in the eye of the beholder?
Conjunctions of meaningfully parallel events (CMPE) are an extreme form of synchronicity that can potentially overcome this subjectivity. The phenomenon requires at least two independent events that occur within 12 hours of each other and that share a long list of common features or parallels.1 This list of parallels needs to tell a coherent story that can then be viewed as commentary on a relevant situation in the life of the person to whom the CMPE happened. The story frames that situation in a certain way and thus conveys a “message” about it.
The specific parameters of this model can determine clearly whether something is a CMPE. The extreme parallelism between the two events makes the phenomenon unusually resistant to a chance explanation. Clear rules of interpretation allow different interpreters to derive the same message from a given CMPE. Here, then, is a form of synchronicity that naturally lends itself to scientific study.
The logical first step in studying this phenomenon is verifying its occurrence. The model was developed by me (Perry) through 30 years of observation and recording of hundreds of examples. However, my examples come from my personal life or the lives of family, friends, and colleagues, not from books on synchronicity and coincidence, which contain only occasional CMPEs. This raises the crucial questions: Is this a genuine phenomenon? If it has not been widely documented, can it really be happening to people everywhere? Should CMPEs be included as a new item on the list of recognized paranormal events, such as near-death experiences and precognition; or are they an idiosyncratic blip?
This independent pilot study was an initial attempt to answer these questions. Its primary goal was to record CMPEs in the lives of its participants.
To recruit participants, we posted announcements on Boston Noetics, an email discussion forum for those interested in the work of the Institute of Noetic Sciences, and on Sign Posts ( www.semeionpress.com), the blog associated with CMPE phenomenon. Additionally, three of us invited several friends and colleagues to participate. This resulted in 17 participants, all sympathetic to and interested in the concept of synchronicity. We asked participants to read my book on CMPEs, Signs: A New Approach to Coincidence, Synchronicity, Guidance, Life Purpose, and God’s Plan, and to attend, or listen to recordings of, three 1-hour-long telephone classes, taught by me and my co-author, Nicola Perry. These focused on understanding the CMPE model, on noticing CMPEs by being vigilant for strikingly similar events, and on performing various activities (such as acting on an important decision) that may help elicit CMPEs.
Once the study period began, we instructed participants to watch their lives closely for any possible CMPEs and to immediately communicate these to me via email. If the examples appeared to pass the initial tests for a CMPE — independence and proximity of events, and number of parallels — I would follow up and gather the information needed to produce a full report, averaging 2,000 words in length. Careful attention was given to participants’ accuracy, including phone conversations, abundant email correspondence, and participant verification of the final report. I then modified this report to mask personal details and sent it out to all participants, as a way to keep interest alive and as further education in the CMPE model.
The study period began on June 1, 2010, and lasted 4 months. It was shortened from the original plan of 6 months because of a larger than expected number of CMPEs, the recording of which became very time-consuming, and because the number of possible CMPEs diminished with each passing month.
During the study period, participants were asked to fill out the Weird Coincidence Scale-2 (WCS-2),2 which measures the prevalence of and beliefs about weird coincidences. We added an additional scale that asked participants which of 13 major life areas they felt a need to change, as changes in these areas are often accompanied by synchronistic events.
Methods of Study
We quantified all possible CMPEs using a scale containing 10 criteria. With each of the 10 criteria, a CMPE of average strength on that criterion would score 5. Thus, any score of at least 50 (average of five for each of the 10 criteria) qualified something as a CMPE. A potential CMPE was disqualified when it failed on any of five key criteria. Here is the scale we used in brief:
Number of events. One event: disqualified; two distinct events: score of 5; three events: score of 7; four events: score of 9.
Independence of events. Events were independent if they “just happened” to occur together, rather than being brought together by normal causation. Clearly dependent: disqualified; very likely dependent: 1; somewhat dependent: 3; a question about independence: 4; clearly, or at least reasonably, independent: 5.
Subject markers. These were characteristics of an event that identified that event as the probable subject being addressed by the CMPE. They amounted to various features that marked an event as personally significant, such as important news, a new idea, an urgent problem, or a sense of emotional significance (see Figure 3, page 581). Neither event had subject markers: 4; one event had them: 5; additional events had them: +1 for each additional event.
Figure 3. Subject markers. Source: Perry R. Reprinted with permission.
Proximity. This concerns how close in time the two events were. More than 12 hours apart: disqualified; 5 to 12 hours apart: 2; 31 minutes to 5 hours apart: 5; 0 to 30 minutes apart: 7.
The next five points concerned the parallels. Because parallels were the features shared by both events, the list of parallels was the measure of the degree of similarity between them. Given how crucial this list is to verifying a CMPE, it is important to understand how it was constructed.
A parallel should be somewhat non-routine (eg, “The man had two legs” would not qualify.) It also should be a prominent part of an event, rather than a peripheral detail. And it should be cohesive — it should fit with the other parallels. Each parallel is validated by the combined strength of these three factors.
Division of parallels is also important, since their number is a key part of the scale. In the CMPE model, parallels are divided into their smallest reasonable units, so that if two elements logically or typically go together, they are left as one parallel. However, if two elements can be pictured apart easily, or are apart in at least one of the events, they are split in two:
Number of parallels. one to three parallels: disqualified; four parallels: 0; five parallels: 3; six to eight parallels: 5; nine to 14 parallels: +0.5 for each one.
Parallels: event-specific. An event-specific parallel was explicit in both events, as something actually said or done, or as a basic description. In contrast, a situational parallel was, in at least one event, a background fact that was closely related to the event itself. Less than 25% of the parallels are event-specific: 0; 25% to 49%: 1; 50% to 74%: 3; at least 75%: 5.
Parallels: specificity. None of the parallels seemed particularly specific or unusual: 2; one or more seemed notably specific or unusual, but not impressively so: 5; one or more of the parallels was impressively specific or unusual: 7.
Parallels: cohesion. The list of parallels needs to tell a coherent story. The parallels were scattered, with only vague themes uniting some: disqualified; most of them were relatively cohesive: 3; all of the parallels were obviously cohesive: 5.
Parallels: centrality. The story told by the parallels is called the “generic situation,” as it was general enough to cover both events. The generic situation was peripheral to both events: 0; it is central to one but peripheral to the other: 3; it was central to each event: 5.
Subject situation. The subject situation was the situation about which the CMPE “was commenting.” It was typically a situation in the person’s life that closely resembled the story told by the parallels (the generic situation) and was usually found in the event containing subject markers. There was no apparent candidate for a subject situation: 0; there was a reasonable but weak candidate: 4; there was a clear subject situation: 5; the generic situation fit a situation of intense present concern for the experiencer or someone he or she is trying to help: 7.
Nicola Perry and I were two of the scorers in the initial study. Our scores cannot be considered truly independent since we discussed the various examples as they were reported. Greg Mackie, a long-time colleague of ours who has extensive experience with CMPEs, also scored all the potential CMPE; his scores can be considered independent. In the future, we will seek the funding to train scorers who will work independently.
We also wanted to compare pilot study CMPEs against CMPEs from my personal records on characteristics such as number of events, proximity of events, and number of parallels. For this purpose, we chose from my records 50 consecutive CMPEs from a period (October 2009 to December 2010) that straddled the study period.
In all, 40 potential CMPEs were collected; 16 were scored as genuine, 23 were scored as failed. One possible CMPE was excluded because, although it appeared to be a genuine CMPE observed by a study participant, it actually happened to someone not part of the study.
The scoring was consistent between scorers. There was disagreement on only two of the 40 examples: one turned on how to define an event, and the other on the quality of the parallels. The average difference between any two scores on a given CMPE was 1.17. The bias factor seemed to have little or no effect as the two closest scorers were Mackie and Nicola Perry (average difference: 0.44), while Nicola and I were much further apart (1.44).
To give a sense of what a CMPE is like, here is an example collected in the study:
Event 1. Alec reads an updated version of Hamlet (10 a.m. to 10:40 a.m.). Alec: “I was reading The New Cambridge Shakespeare version of Hamlet, edited by Philip Edwards, and found myself consciously reflecting on Edwards’ footnotes. They are very detailed, and the level of detail drew my attention. My first response was to be amused by the detail — for instance, the pains he takes to explain what a particular word means, and to chase down other instances of the same word being used in the Shakespeare canon. ... On further reflection, I conceded that, actually, the thoroughness of the footnotes was a good thing. I was finding them useful, and they allowed me to feel confident about Edwards as editor.
“A blurb near the front of the book says, ‘Philip Edwards aims to bring the reader, playgoer, and director of Hamlet into the closest possible contact with Shakespeare’s most famous and perplexing play.’”
Event 2. Alec opens and replies to email (10:40 a.m.). Alec: “I opened and immediately replied to an email from my friend Michael. In it, he specifically asked me my view on his footnotes in his new translation of the Gospel of Thomas.
“Michael is doing a reconstruction of the Gospel of Thomas, trying to restore it to its original wording. This means deciding between the Coptic version and the Greek fragments, as well as filling in lacunae. His aim is to give the reader a sense of being in direct contact with the original gospel.
“Michael, however, had become concerned over his extensive footnotes. One kind of note explains his editing decisions and the other kind explains some of the gospel’s difficult language. He worried that these might be intrusive for the reader. In this email, he asked whether the footnotes, with their explanations of passages, ‘go too far? They are actually the part I’m most worried about.’”
Proximity: Simultaneous — the two events met at 10:40 a.m.
Parallels: All of the parallels below are found in both events, even though this may not be apparent from our shortened account:
Alec is or has been reading an important piece of literature.
Multiple versions of this work exist.
He is specifically reading a newer edit of this work done by a scholar.
The point of this edit is to bring the reader into the closest possible contact with the original work.
This is very difficult, in part because of scribal errors.
The scholar has (or intends to) put cross-references in his work.
He has also extensively footnoted his version.
One kind of note is there to explain and justify his editing decisions.
Another kind is to act as a bridge between the reader and the text, helping the reader understand the difficult language.
Alec reflects on or is asked about the appropriateness of such extensive footnotes.
All three scorers scored this CMPE at 57.5. There were two distinct events (score of 5 on criterion #1) that were, as far as we (or Alec) could tell, completely independent (5 on #2). The second event had a subject marker: a request (5 on #3). The events were simultaneous (7 on #4). They share a long list of parallels (6.5 on #5). (The original number of parallels was 11, but to protect privacy, we masked some personal details, which required omitting one of the parallels.) All of the parallels, with the exception of #6, were event-specific, in that they were either explicitly said or done in both events or constitute a basic description of the events (5 on #6). The parallels are impressively specific, especially parallels 5 through 10 (7 on #7). They tell a clearly cohesive story (5 on #8) that is central to both events (5 on #9). Finally, this story fits a situation of intense present concern for someone Alec is trying to help (7 on #10).
In evaluating the data, we compared the examples that scored as genuine with those that failed. The average score for the 16 genuine CMPEs was 55.65. In contrast, 22 of the 23 failed CMPEs received a score of 0, in that they were all disqualified for one or more reasons, most of them (17) because of an insufficient number of parallels. Only one of the failed CMPE had a score above 0, at 45.7.
Figure 1 (see page 580) compares the number of events in the study’s 16 CMPEs and the 50 personal CMPEs from my records, expressed as a percentage of the total.
Figure 1. Number of events in the study’s 16 CMPEs and the 50 personal CMPEs from Perry’s records, expressed as a percentage of the total. Source: Perry R. Reprinted with permission.
Figure 2 (see page 580) compares the kinds of events in the study CMPEs with my collected CMPEs, each expressed as a percentage of the total number of events. CMPEs are typically rich in information, ideas, or story elements.
Figure 2. Comparison of the kinds of events in the study CMPEs with Perry’s collected CMPEs, each expressed as a percentage of the total number of events. Source: Perry R. Reprinted with permission.
Figure 3 compares the presence of “subject markers” in the events of the study CMPEs with my collected CMPEs. These are the features of an event that mark it as the probable subject of the CMPE. Each of these is expressed as a percentage of the total number of events.
Figure 4 compares the proximity of the events in the study CMPEs and my collected CMPEs. Based on past tracking, I determined that proximity tends to fall into three zones: 55% of CMPEs in the first zone (events were within 0 to 30 minutes of each other); 35% in the second (31 minutes to 5 hours); and 10% in the third (5 to 12 hours). The results below approximate that same distribution.
Figure 4. Proximity of events. Source: Perry R. Reprinted with permission.
Figure 5 (see page 582) compares the number of parallels in the study’s genuine CMPEs with those that failed (those that made it as far as the parallels, having not been disqualified earlier). The failed CMPEs averaged 2.18 parallels, compared with an average of 9.75 for the genuine CMPEs; 50 of my collected CMPEs averaged 7.94 parallels. In designing the scale for the study, I measured my CMPEs during a 3-year period and found an average of 9.03.
Figure 5. Number of parallels. Source: Perry R. Reprinted with permission.
Figure 6 (see page 582) compares the subject situations in the study CMPEs and my CMPEs. The subject situation is what the CMPE appears to be about. This is, of course, a matter of interpretation, but as explained in criterion number 10 of the scale, there are rules behind the interpreting. Each of these is expressed as a percentage of the total CMPEs.
Figure 6. Subject situations. Source: Perry R. Reprinted with permission.
Consistent with earlier findings, CMPEs are often interrelated. This observation was supported by the study: Three of the four participants who had multiple CMPEs had CMPEs that were obviously related to each other. For one participant with three CMPEs, all of them concerned the purpose of a recent move. For another with five CMPEs, all of them included the meeting place of psychological trauma and spiritual transcendence, a major theme in her life and work.
The 16 CMPEs were dominated by high-frequency scorers, something also expected based on previous observations: one participant had five; two had three each, one had two; three more had one apiece; 10 had none. The top four scorers thus accounted for 81% of the genuine CMPEs. We do not know, however, if these people actually had more or just noticed more, as the same top four scorers all reported failed CMPEs, accounting for 43% of the failures.
All of our participants took the WCS-2. Scores for this survey range from insensitive, to average, to sensitive, to very sensitive, to ultrasensitive. The scores of those who reported CMPEs (37.71) and those who didn’t (37.9) were almost exactly the same. Despite this, high-CMPE scorers tended to cluster at the top of the WCS-2 scale (our top three scorers were either ultrasensitive or very sensitive); the three who scored “insensitive” on the WCS-2 all reported no CMPEs. However, the correlation between the patterns of CMPE success and those of WCS-2 scores did not attain statistical significance.
As an addition to the WCS-2, participants were asked what areas of their lives, of 13 major areas, they felt a need to change. This, however, predicted only 17% of the subjects that were actually addressed by CMPEs. A much more reliable predictor was asking about the areas in which they had experienced key events (eg, births, deaths, marriages, accidents, new relationships, new job) in the past 3 months. This predicted 44.7% of the actual CMPE subject situations.
Overall, participants experienced genuine CMPEs as measured by our scale, with notable frequency. In every respect, these CMPEs seem to be the exact same phenomenon that I have described. This was the case according to the scoring scales, and was also supported by comparison to the CMPEs I personally recorded.
These CMPEs were evaluated by the scale as markedly different from the failed examples. The two groups were not scattered along a single continuum, but rather largely fell into two widely separated classes — completely disqualified and clearly qualified. There was only one exception to this, the failed CMPE that was initially disagreed on but was eventually scored at 45.7.
Interestingly, there were six CMPEs which happened to me before and after the study and concerned the study itself. Three emphasized the importance of measuring possible CMPEs using the scale, and one suggested we showcase the difference between the failed and genuine CMPEs, especially in terms of number of parallels. Others were quite optimistic about the study’s outcomes. The sixth CMPE occurred as I finished the draft of this paper. Immediately after, I read a published story of Dr. Andrew Newberg’s initial study on measurable correlations between brain states and meditative experience.3 Newberg’s study, which contains many parallels to the CMPE pilot study, showed for the first time that the ancient phenomenon of contemplative experience has a scientifically measurable dimension and, thus, sparked further research in this area.
These CMPEs, then, both gave us a positive framework through which to view the study and guidance for how to do it (“Emphasize the quantified measurement of the phenomenon”). This highlights one of the important benefits of CMPEs: They give us a frame of reference from which to give meaning to the events of our lives, meaning that can be beneficial on both a psychological and practical level.
Both of our major results — the recording of CMPEs and the gulf between them and failed examples — are suggestive of CMPEs being a distinct phenomenon within the spectrum of synchronicity and coincidence. This pilot study, then, has provided promising initial support for what in time could prove to be the most testable form of synchronicity and an important paranormal phenomenon.
- Perry R. Signs: A New Approach to Coincidence, Synchronicity, Guidance, Life Purpose, and God’s Plan. Sedona, AZ: Semeion Press; 2009.
- Coleman SL, Beitman BD, Celebi E. Weird coincidences commonly occur. Psychiatr Ann. 2009;39(5):265–270. doi:10.3928/00485713-20090421-03 [CrossRef]
- Volk S. Fringe-ology: How I Tried to Explain Away the Unexplainable—and Couldn’t. New York: HarperOne; 2011:177–182.