Is String Theory the Ultimate Theory of Everything?

[Chronos]

String theory has neither been confirmed, or disproven by experiments to date. Its premises are difficult to constrain and more than a few believe the relevant energy levels are far beyond our present or future technological capabilities. That suggests the only hope is astrophysics. The universe is capable of inconceivable energy levels. The trick is where and what to look for.

 

[arivero]

Is string theory going in a wrong direction? Is it going anywhere, in any case?

 

[ChrisVer]

What do you mean by "in a wrong direction"?
The string theory as it is, is a mathematical theory. It still lacks the mathematics needed to be completed and that's what people are working on. They are roughly speaking not physicists neither mathematicians but something hybrid...
It has so many available windows, that IMO one of them will have to be correct at some point...
At the moment, the only way it's going (as I see it) is building up the new mathematics it needs.

 

[nuclearhead]

Is string theory going in a wrong direction? Is it going anywhere, in any case?

Maybe. How would we know? We'll only know which direction is right once it is complete.

And by complete it has to:

• Be able to make calculations at high energies and small distances.
• Be able to predict the masses, charges and spins of all yet undiscovered particles.
• Be able to give correct answers near intense gravitational fields like a black hole.
• Not give answers as series that diverge.
• Satisfy all known symmetries and some that are unknown.
• Be consistent when applied to the whole Universe.

 

[ohwilleke]

I've heard it said that SUSY and SUGRA are necessarily low energy effective theories in any String theory. If this is the case, then it might very well be possible to rule out SUSY. We have a variety of LHC and other exclusions that push up the minimum mass scale of any SUSY/SUGRA theory in order to avoid discernable differences from the SM in experiments so far.

It only takes one decent, not very strong exclusion in the other direction from phenomenological properties of very high energy scale SUSY/SUGRA theories that manifest at low energies (e.g. neutrinoless beta decay rates) to create a vice that excludes 100% of the SUSY/SUGRA parameter space. If ruling out 100% of the SUSY/SUGRA parameter space suffices to rule out String theory, then this is something that could happen in our lifetimes (at least for the younger participants in this forum).

Another reason to take pause at String theory is that it goes to really great lengths, for example, requiring 7 extra compactified dimensions, to integrate gravity and the other SM forces (which are often confined to a four dimensional manifold). If you have to contort the overall theory so much simply to secure that one feature, maybe you are taking the wrong approach to quantum gravity.

I wouldn't be surprised if some of the insights of string theory are pertinent to a deeper theory that explains the SM and GR's loose ends better, but now that the Higgs boson mass insures that the SM doesn't break down anywhere all of the way up to the Planck scale, the absolute need for it is much less compelling.

 

[arivero]

What do you mean by "in a wrong direction"?

At the moment, the only way it's going (as I see it) is building up the new mathematics it needs.

Maybe. How would we know? We'll only know which direction is right once it is complete.

Well, if QFT were mainly used by their developers in a field foreigh to particle physics, say prediction of stock market, I'd say that it is not wrong, but it is going in a wrong direction. And of course arguments of mathematical learning and internal consistency could be given to explain that all the development and discoveries we do while looking to the renormalisation group of stock options (I am just joining words here, not claiming that such thing does exist) will allow to find the right formulation of the theory and then understand its use in particle physics. Yes, could be. And you can go to Japan from Europe traveling westwards. But it is the wrong direction.

 

[arivero]

I've heard it said that SUSY and SUGRA are necessarily low energy effective theories in any String theory.

...

Another reason to take pause at String theory is that it goes to really great lengths, for example, requiring 7 extra compactified dimensions,

...

Nice, such are the reasons that make me to think that string theory is right and that it just happens not to be going in the right direction, namely to try to fit with the spectrum of HEP particles at the QCD-electroweak energy level. Which is the thing for which String Theory was invented.

The point that the smallest spaces whose isometry group is SU(3)xSU(2)xU(1) happen to be 7-dimensional makes me to believe that the M-theory idea could have some fundation. Furthermore, the number of degrees of freerom of the "low energy M-Theory", aka supergravity, is the same that in the SUSY SM with right neutrinos. I even imagine that the need of an 8th dimension to build manifolds with Pati-Salam symmetry is related to the existence of F-Theory or other string-theoretical.

As for SUSY itself, remember that my own pet theory looks very much as having an open string terminated in five different flavours. So I had hoped that a string theory worried about the GeV scale could also find this scheme. It could have happened if they had followed some early papers of Schwarz in 1971, where he considered both quarks and gluonic strings as part of the same scheme. Or could have happened afted the discovery of SUSY with some variant of the bootstrap ideas of Chew; in the original bootstrap you ask every particle to be a composite of all the others; with SUSY you can just ask the susy partners to be composite of the known elementary objects, and then it automagically implies three generations.

 

[arivero]

The limits for mass particles don't come from Supersymmetry I guess, they come from the models of Supersymmetry. So models can of course be wrong-
I remembered the value- when the top quark was proposed (right after the discovery of bottom), the models we made proposed that it should have a mass of roughly 17GeV... they didn't find it there, so they said it must be around 30, 40, 50GeV the most... The top quark appeared at 170GeV, and from that we get some lesson. Firstly, the nature will show itself up as it is and not as we want it to be. And secondly, we don't know why the top quark must be so heavy.

For instance, a guy in that time, using SUSY and a Chew-like Bootstrap idea, had been able to predict the third generation... and a heavy top. This is discussed in old threads; let me to sketch the argument.

1) Demand that [scalar] superparticles are composites of [fermionic] particles; this is, that the number of degrees of freedom must be the same, for each sector of charge. So N flavours UP and M flavours DOWN will bind to form sparticles, "diquarks" if you wish, under a SU(N)xSU(M). One finds that there is no non-trivial solution... but...

2) Demand that only "Light" quarks can bind to build such composites. Then you have a series of solution, with "heavy" families, "light" families and perhaps a "mixed" one. The simplest solution of the series has two "light" and one "mixed", the heavy quark being of charge +2/3.

3) [bonus] Use quark-antiquark to build the superpartners of leptons. It fits perfectly with the number of charged leptons (it is the same condition than for down-type quarks) and predicts 12 neutral scalar degrees of freedom, so it predicts right neutrinos too.

Such argument could have happened in the seventies. But the superstring & susy guys had already buried the bootstrap and started to sail towards Planck Scale, the gold pot at the end of the rainbow. Had they being able to predict the third generation and its peculiar light-heavy mix, they had shown that "string methodology" can do predictions.

 

[nuclearhead]

For instance, a guy in that time, using SUSY and a Chew-like Bootstrap idea, had been able to predict the third generation... and a heavy top. This is discussed in old threads; let me to sketch the argument.

1) Demand that [scalar] superparticles are composites of [fermionic] particles; this is, that the number of degrees of freedom must be the same, for each sector of charge. So N flavours UP and M flavours DOWN will bind to form sparticles, "diquarks" if you wish, under a SU(N)xSU(M). One finds that there is no non-trivial solution... but...

2) Demand that only "Light" quarks can bind to build such composites. Then you have a series of solution, with "heavy" families, "light" families and perhaps a "mixed" one. The simplest solution of the series has two "light" and one "mixed", the heavy quark being of charge +2/3.

3) [bonus] Use quark-antiquark to build the superpartners of leptons. It fits perfectly with the number of charged leptons (it is the same condition than for down-type quarks) and predicts 12 neutral scalar degrees of freedom, so it predicts right neutrinos too.

Such argument could have happened in the seventies. But the superstring & susy guys had already buried the bootstrap and started to sail towards Planck Scale, the gold pot at the end of the rainbow. Had they being able to predict the third generation and its peculiar light-heavy mix, they had shown that "string methodology" can do predictions.

When Einstein tried to create his unified theory, of gravity and electromagnetism it didn't work because a few years later more particles and forces were discovered.

Any theory today that predicts only the Standard Model particles will almost certainly be wrong as more particles and forces are discovered in the future.

String theory, at least, says that no matter how much energy you put into a machine like the LHC you will always find new particles of higher and higher masses. (The modes of a string).

To me this seems more likely than to say (like Einstein mistakenly did) "we've discovered everything - all we have to do is find the pattern". Or, like the captain of the Titanic, "that Iceberg doesn't look very big" when 99% of it is under the water.

So if you think that there will always be new particles at higher energies then you have to either believe in string theory or kalulza-klein theories as they are the ones that predict this. But this is impossible to prove experimentally, so you can only prove it with math and show that the infinities go away if you accept these heavy particles exist.

 

[arivero]

So if you think that there will always be new particles at higher energies then you have to either believe in string theory or kalulza-klein theories as they are the ones that predict this. But this is impossible to prove experimentally, so you can only prove it with math and show that the infinities go away if you accept these heavy particles exist.

So you also think that the String scale is not Planck scale, but TeV or near? Huge extra dimensions?

 

[cosmik debris]

Now that's just silly. No one area of study precludes any other area of study.

Not intellectually, but financially maybe?

 

[ohwilleke]

Any theory today that predicts only the Standard Model particles will almost certainly be wrong as more particles and forces are discovered in the future. . . . So if you think that there will always be new particles at higher energies then you have to either believe in string theory or kalulza-klein theories as they are the ones that predict this. But this is impossible to prove experimentally, so you can only prove it with math and show that the infinities go away if you accept these heavy particles exist.

FWIW, I am not at all certain that more particles and forces will be discovered in the future. The SM particles plus a graviton plus a better understanding of the fabric of space-time may be it. If I were a Baysean, I might assign a prior probability of 33% to a scenario like that.

* Even if there are more particles, the number may be quite few. We may need new physics, and with it, new particles, to explain dark matter, dark energy, inflation, the strong CP problem, neutrino mass and oscillation, and baryogenesis and leptogenesis. But, we might not.

If there are dark matter fermions, there could be a set as complex as the SM fermions, there could be a triplet of sterile neutrino-like particles, or there could be a spin-1/2 or spin-3/2 singlet gravitino.

There might be in the dark sector, a set of bosons as complex as the SM bosons, there could be a DM self-interaction boson (perhaps a MeV scale mass dark photon), dark matter itself could be an axion-like boson (which might also give rise to dark energy), there might be a dark energy scalar boson, or there might be a couple of extra graviton-like bosons (the spin-2 graviton we know and love, a spin-1 vector graviton to give rise to DM effects, and a spin-0 graviton to give rise to dark energy).

There might be a separate inflation boson (probably a scalar or a tensor), or inflation might arise from some source that also explains something else (e.g. the Higgs field or a unified GUT boson or the fabric of space-time's properties or gravitational potential energy in a near singularity regime).

There might be a light Z boson-like particle that facilitates neutrino oscillation (perhaps also interacting with dark matter to generate neutrino mass in a see-saw mechanism) or there might not.

There might be an axion that addresses the strong CP problem and has no role in the dark sector or inflation or neutrino oscillation, but I really doubt it.

There might be a heavy W/Z boson-like particle that violates B and L number conservation or the universe's non-zero B and L number might have no discernible particle or force mechanism to explain it any more than we have a particle or force mechanism to explain why the universe has precisely the amount of mass-energy that it does.

* Honestly, I really doubt that there are even that many. My money would be on 5-6 more at most, and probably less. Dark energy, inflation and B and L violation, are probably more likely to be manifestations of already known particles and forces that behave in unexpected ways under certain circumstances.

For example, an inflation may be the form the SM gauge bosons (i.e. the photon, gluon, W and Z bosons) take at the GUT scale when there is a gauge unification, or might be a high energy manifestation of the Higgs boson and field, or it might arise from quantum gravity effects. Dark energy might very well be a manifestation of a baseline of graviton or photon or both kinds of radiation or a property of space-time itself. B and L might not be violated at all and have had their values since the Big Bang and not be broken symmetries

The DM sector might very well not have its own boson and only have a DM fermion. A singlet DM particle is as likely as a triplet in my mind. Then again, DM may not exist at all with its phenomena actually due to quantum gravity effects, or the exclusion of gravitational waves with lengths approaching the length of the universe, or ill appreciated non-Newtonian aspects of GR, or due to one or two more gravitational bosons in addition to the tensor graviton (e.g. perhaps a vector graviton giving rise to DM phenomena and a scalar graviton giving rise to dark energy).

At this point, I think that the likelihood of a zoo full of new fundamental particles a la the extra Higgs bosons and superpartners of SUSY is remote (probably less than 5%).

* I could imagine, and at some level expect, that all "fundamental" particles are actually made of some smaller subset of preons bound together by a single preon binding force with a boson to carry it, or even from just one or two kinds of strings. While, I suppose that this kind of substructure would qualify as new particles or forces, substructure like this might very well manifest only in the existing particle set.

* Also, I could imagine that we actually have too many particles already.

For example, I could imagine that the Higgs boson is really some manner of composite of the photon and the W+, W-, and Z bosons (e.g. if the photon and Z have opposite spin, and the W+ and W- have opposite spin, these four have a combined spin-0 even like the SM Higgs, and have the same combined charge, further the Higgs boson mass is very close to the masses of these four electro-weak bosons combined divided by the square root of four (the number of bosons being combined), and all particles that have a Higgs boson Yukawa also interact with the W and Z bosons, unlike the gluon which does not).

* I could also imagine that the existing SM forces unify and become indistinguishable from each other at a GUT scale, giving rise to a unified GUT boson that might be critical in understanding inflation, baryongenesis, leptogenesis and DM creation, but I wouldn't necessarily call that a new particle or force.

* I could also imagine that there are composite particles that exist in some circumstances but have not yet been discovered (e.g. various kinds of lepton-lepton atoms like muonium, unstable baryons that are stable in esoteric high energy conditions and form atoms in those circumstances, rare and short lived top quark hadrons, tetra/penta/hexa/septaquarks, glueballs and glue-quark hybrid particles).

* Similarly, I could imagine new effective "spillover forces" like the nuclear binding force between baryons which is a spillover of the strong force and carried by pions.

 

[ChrisVer]

For instance, a guy in that time, using SUSY and a Chew-like Bootstrap idea, had been able to predict the third generation... and a heavy top. This is discussed in old threads; let me to sketch the argument.

1) Demand that [scalar] superparticles are composites of [fermionic] particles; this is, that the number of degrees of freedom must be the same, for each sector of charge. So N flavours UP and M flavours DOWN will bind to form sparticles, "diquarks" if you wish, under a SU(N)xSU(M). One finds that there is no non-trivial solution... but...

2) Demand that only "Light" quarks can bind to build such composites. Then you have a series of solution, with "heavy" families, "light" families and perhaps a "mixed" one. The simplest solution of the series has two "light" and one "mixed", the heavy quark being of charge +2/3.

3) [bonus] Use quark-antiquark to build the superpartners of leptons. It fits perfectly with the number of charged leptons (it is the same condition than for down-type quarks) and predicts 12 neutral scalar degrees of freedom, so it predicts right neutrinos too.

Such argument could have happened in the seventies. But the superstring & susy guys had already buried the bootstrap and started to sail towards Planck Scale, the gold pot at the end of the rainbow. Had they being able to predict the third generation and its peculiar light-heavy mix, they had shown that "string methodology" can do predictions.

I don't really understand what you want to say by that. Also I am not familiar of this theory [why would someone use fundamental particles to make a fundamental particle?]. What I know for sure however, is that the way the extra dimensions of the theory are compactified gives the information about the particles you expect to have in your model. The way this compactification occurs however is pretty non-trivial, and it's very difficult to study its geometry.
Also the string theory is not a complete theory. It has several "edges" all of which can be connected to M-Theory which is unknown, because we lack the mathematics for that.

 

[arivero]

I don't really understand what you want to say by that.

Well, it was just a anachronistic example, of other path that string theory could have taken in the early seventies.

Yep, nobody is familiar with bootstrap nowadays, and for sure I am not. I understand that it was a set of loose ideas from Chew, the tutor of David Gross if I recall correctly, and that they motivated the study of the S-Matrix looking for some self-consistent solucion, so the name "bootstrap". It is the historic origin of string theory.

The way this compactification occurs however is pretty non-trivial, and it's very difficult to study its geometry.

Here again, my opinion is that there are other directions in compactification. The groupthinking of string theoretists drove them to look for gauge groups first from SO(32) and E8, and later this path was disfavoured and interesecting branes were proposed, but they never took seriously Witten 1981 compactification; in part because Salam et al were unable to get the right spectrum of leptons and fermions, but mostly because effort was diverted elsewhere.

 

[Sousf]

We have found no evidence whatsoever to support string theory and supersymmetry and to prove that the theory is correct.
Although the "Large Hadron Collider" has found some evidence to support the Higgs boson particle(I think they found a trace of its energy)there would still be along way to proof that the theory is right.But still,if they can find the Higgs particle it is going to be a very big thing in science.

 

[ChrisVer]

We have found no evidence whatsoever to support string theory and supersymmetry and to prove that the theory is correct.
Although the "Large Hadron Collider" has found some evidence to support the Higgs boson particle(I think they found a trace of its energy)there would still be along way to proof that the theory is right.But still,if they can find the Higgs particle it is going to be a very big thing in science.

The Higgs is "long"-found (and nobel prize won)...

 

[Sousf]

Oh,ok

 

[ChrisVer]

just to give a complete answer...
The Higgs was found, what is missing is a better significance (larger than 5 sigma) to say [experimentally] that it is found. However most of the scientists have concluded that it's found,the certainty is so large that they handed in the Nobel Prize to Higgs and Englert last year...The new run of LHC is going to get more data around Higgs and thus finish the job.

 

[Sousf]

Ok,because all the article I read only said they only found part of it but what do I'm only 14

 

[ChrisVer]

"part of it" : well you can't find "part of" Higgs. Probably you are trying to say that we have found 1 Higgs, whereas some other theories predict more Higgses, then yes that's true.
Otherwise your articles were just trying to point out the fact that we need a better signal (having greater than 5 sigma significance, and as a result of this, minimizing the statistical errors significantly).

 

[Sousf]

Ok thx

 

[ohwilleke]

The current significance of the Higgs boson discovery is currently in the vicinity of 6.8-8 sigma, well past the discovery threshold.

More importantly for the string theory discussion, SUSY theory generically predicts at least five spin-zero Higgs bosons (a light and heavy neutral scalar boson, a neutral pseduoscalar boson, and a positively and negatively charged Higgs bosons) and some predict additional sets of four. To the extent that string theory has SUSY as a low energy effective theory, the non-discovery of extra Higgs bosons narrows the parameter space and disfavors the entire theory.

 

[ChrisVer]

when did it reach 6.8-8 sigma?
Also I think it's better to replace SUSY with MSSM.

 

[ohwilleke]

when did it reach 6.8-8 sigma?

This summer.

 

[PAllen]

Though supersymmetry is but one aspect of string theory, I think the following is a really honest description of the current state of affairs, by a SUSY enthusiast of great stature:

http://arxiv.org/abs/1309.0528

While this author makes a compelling case that it is wholly premature to rule out SUSY, he makes the following admission (reflecting the need to give up either naturalness or parsimony):

"Candor compels me to declare that at this time there is no supersymmetry as
our forebears understood the term, and as they meant it to be understood by
us."

 

[atyy]

Here are some discussions of the Higgs mass.

http://arxiv.org/abs/1205.2893
Higgs boson mass and new physics
Fedor Bezrukov, Mikhail Yu. Kalmykov, Bernd A. Kniehl, Mikhail Shaposhnikov

http://arxiv.org/abs/1307.3536
Investigating the near-criticality of the Higgs boson
Dario Buttazzo, Giuseppe Degrassi, Pier Paolo Giardino, Gian F. Giudice, Filippo Sala, Alberto Salvio, Alessandro Strumia

Here are some proposals about how string theory may be consistent with the observed Higgs mass.

http://arxiv.org/abs/1206.2655
The Intermediate Scale MSSM, the Higgs Mass and F-theory Unification
Luis E. Ibáñez, Fernando Marchesano, Diego Regalado, Irene Valenzuela

http://arxiv.org/abs/1301.5167
The Higgs Mass as a Signature of Heavy SUSY
Luis E. Ibanez, Irene Valenzuela

http://arxiv.org/abs/1304.2767
The Higgs mass from a String-Theoretic Perspective
Arthur Hebecker, Alexander K. Knochel, Timo Weigand

http://arxiv.org/abs/1406.6071
Towards the Standard Model in F-theory
Ling Lin, Timo Weigand

 

[arivero]

Also I think it's better to replace SUSY with MSSM.

I think that it is more minimal to say SUSY SM, or SSM. The MSSM is one of the extensions that solves the problem of giving mass to both quark sectors. But the really minimal thing if one wants massive W and Z in supermultiplets only adds three scalars, to complete each gauge supermultiplet.

 

[ChrisVer]

if one wants massive W and Z in supermultiplets only adds three scalars, to complete each gauge supermultiplet.

do you have a reference about this procedure?
And I didn't really understand the "scalar" things. You mean chiral sfields?
Of course it doesn't sound fun to avoid adding masses to the quarks so I guess it's just a toy model. Also it won't contain the Higgs sector, and thus it's even more unphysical [I guess it's explicitly breaking the SM symmetries].
Also my interaction with MSSM was that we just took the SM and extended it to supersymmetry (each field → chiral sfield). So I didn't see this distinction to other models where you don't extend everything. Then any other extension, was NMSSM like...

 

[arivero]

do you have a reference about this procedure?.

Do you have any book on supersymmetry?

Well, for instance, Terning's. In the page 9, first line: "The massive vector multiplet... which corresponds to two Majorana fermions, a massive vector (spin 1) and a real scalar".

Of course, the massless vector multiplet has just the massive vector and one majorana fermion, the gaugino.

But if you want to have supersymmetry AND massive gauge fields, you must add to each gauge supermultiplet another Majorana fermion and a real scalar.

So any extension of the standard model where we restore susy but we keep the SU(2)xU(1) symmetry broken, which we could in principle do if susy breaking is independent of electroweak symmetry breaking, should have at least three extra real scalars, for the multiplets of Z0, W+ and W-.


EDIT: I am not sure if it contains or not the Higgs sector, but I guess that any Higgs sector will just include this three scalars, as surely when the massive gauge multiplet becomes massless it will produce a separate chiral multiplet with the extra Majorana, the extra scalar, and other scalar corresponding to the degree of freedom "eaten" by the massive spin 1 vector field. So it looks to me as a higgs.

I would be surprised if someone could exhibit a model extending the SM and not containing this three scalars, and their corresponding chiral supermultiplets. Because of it, I say that this "SSM" is the most minimal. The only way to avoid is is to build a model where susy breaking and electroweak breaking are retorted in a way such that you can not manipulate the langrangian parameters to restore susy and keep at the same time the electroweak bosons massive.

 

[ohwilleke]

I would emphasize, instead, the very narrow parameter space of two Higgs doublet models of any kind subject to very general and generic assumptions that remains given current experimental data. See http://arxiv.org/abs/1409.3199

If the two Higgs doublet parameter space is closed, then you either have the SM's one Higgs doublet (which rules out supersymmetry) or very non-minimal SUSY models within nine or more spin-0 Higgs-like particles (including charged and double charged Higgs bosons), which add an immense number of particles, forces and parameters, none of which has any experimental support whatsoever so far.

 

[KL7AJ]

I don't think we'll be putting String Theory in a test tube any time soon. Unfortunately any physicist who doesn't swallow String Theory is pretty much a pariah.

 

[Chronos]

IMO, most modern physicists tend to consider ST a pill too big to swallow. It is mathematically elegant, but, since when has that ever been considered an acceptable substitute for empirical evidence?

 

[ChrisVer]

IMO, most modern physicists tend to consider ST a pill too big to swallow. It is mathematically elegant, but, since when has that ever been considered an acceptable substitute for empirical evidence?

Since never. Who said that string theorists are looking for substitutes? Theoretically it has what it needs as you pointed out: it's elegant. Practically it's unproven (as most models beyond standard model). Nothing but it's nice content makes it strong, and in a previous post I wrote that it's main issue at the moment is advancing mathematics rather than physics. IMO if someone calls the string theory a physical theory which explains everything, then either he doesn't understand what physical stands for or what string theory is... (un/)fortunately.

 

[Ilja]

String theory at this point is not right OR wrong, it is simply a hypothesis that has no experimental evidence but which would explain a lot of stuff very nicely if it DOES turn out to describe reality.

Why would you want to abandon the search for a theory of everything? Do you not care about knowledge?

It should not be abandoned. But something should be done to reduce it to what it deserves. Namely, to the status of "wild speculations beyond the standard model".

This is more a problem of organization of science: Scientific bureaucracy should decide how many money will be give to the whole domain "speculations beyond the standard model", and these money should not be specified in any further way. Those who work in this domain should not be considered on equal foot with physicists connected with experiments, but somehow between physicists and philosophers and mathematicians, that means, with much more freedom of choice of the direction they want to work on.

 

[Ilja]

Now that's just silly. No one area of study precludes any other area of study.

In the modern, stupid way of organization of science this is a possibility.

The scientist has to be independent. What one needs for independence is quite clear: Security, that means, a safe job. You have an exceptional idea, which you want to develop yourself, but you need many years for this? No problem if your job is safe. But an big problem if your job is not safe. And an unsolvable one if the only safe thing about your job is that after two years you have to look for another one because the grant is finished.

With this stupid way of organizing science, all scientists without tenured jobs have to care about their next position. That means, to care about publications, citations, conference participations and talks, and all this. And all this is far easier if you work in the latest vogue of the modern mainstream science. And in such a situation one area of study can easily preclude a lot of other areas of study.

It doesn't even matter much what is the latest fashion - the unfashionable research directions are automatically precluded, reduced to a few individuals with tenured positions and independent scientists who do not have to care about jobs.